{"meta":{"page":1,"per_page":50,"max_per_page":100,"total":1145,"total_is_capped":false,"direct_labels_cover":2,"predictions_cover":1145,"direct_label_status":"direct model label, unvalidated","prediction_status":"machine_predicted_unvalidated (Codex and Gemma teacher distillation)","score_status":"score_only:v0-immature-baseline (scores rank; they never assert a category)","snapshot":{"source":"OpenAlex, pinned release, all 482 partitions","release":"2026-06-24","frame_built":"2026-07-12"},"query_hash":"29852c204d50","filters":{"topic":"Reinforcement Learning in Robotics"}},"results":[{"id":"W2754517384","doi":"10.1609/aaai.v32i1.11694","title":"Deep Reinforcement Learning That Matters","year":2018,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":1504,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"Open Philanthropy Project; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Computer science; Standardization; Benchmark (surveying); Artificial intelligence; Field (mathematics); Variance (accounting); Deep learning; Machine learning; Data science; Risk analysis (engineering); Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.07088540185984533,"gpt":0.288866159836522,"spread":0.2179807579766767,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005631779,0.0002654556,0.0002458711,0.0001622093,0.0004062697,0.0004135727,0.002704812,0.00009179958,0.0002288619],"category_scores_gemma":[0.0003217166,0.0002044874,0.0001248941,0.0006154813,0.0004730623,0.0005748551,0.0007581514,0.0004276233,0.000534609],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007581721,"about_ca_system_score_gemma":0.0000639686,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002473288,"about_ca_topic_score_gemma":0.000002897337,"domain_scores_codex":[0.997621,0.00002233459,0.0005475839,0.0004922792,0.000811318,0.0005054875],"domain_scores_gemma":[0.99818,0.00008579525,0.0005699989,0.0004080477,0.0006502195,0.000105944],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005134815,0.00004301838,0.0006511355,0.00004495347,0.00003394356,5.729058e-7,0.003763529,0.06442562,0.01257456,0.8939208,0.0004977569,0.02399278],"study_design_scores_gemma":[0.0000258388,0.0003895584,0.0001051631,0.0001564249,0.000009488826,0.000003856035,0.0005007188,0.7046347,0.2818424,0.01142221,0.0006802078,0.0002293482],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00676992,0.000006312319,0.9510738,0.005290208,0.0007795903,0.0003790251,1.112205e-7,0.0001867467,0.03551434],"genre_scores_gemma":[0.9927465,0.00002035039,0.004614619,0.0009793142,0.0001076901,0.00001597491,3.742789e-7,0.00001667176,0.001498479],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9859766,"threshold_uncertainty_score":0.8338752,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3100366369","doi":"10.1561/2200000071","title":"An Introduction to Deep Reinforcement Learning","year":2018,"lang":"en","type":"article","venue":"Foundations and Trends® in Machine Learning","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":1251,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Artificial intelligence; Computer science; Deep learning; Generalization; Field (mathematics); Robotics; Machine learning; Robot; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.01379196626407731,"gpt":0.2905965508752665,"spread":0.2768045846111892,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006109281,0.0001825817,0.0001655456,0.0007663849,0.0006742777,0.0004633143,0.0003598064,0.0000559686,0.0003595039],"category_scores_gemma":[0.0001992523,0.0001885887,0.00003091576,0.001161934,0.00006312553,0.0008634281,0.0002209635,0.0004805419,0.0001171342],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008844589,"about_ca_system_score_gemma":0.00001712027,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001537651,"about_ca_topic_score_gemma":0.00016753,"domain_scores_codex":[0.9983198,0.0001766471,0.0003470065,0.0005224819,0.0002672375,0.0003668279],"domain_scores_gemma":[0.9991729,0.00005394618,0.0001335348,0.000395868,0.0001038068,0.000139907],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000008118957,0.0000164881,0.009627651,0.000003602339,0.000009371041,0.00000180447,0.001973486,0.8689428,0.0001832247,0.01633217,0.00004572627,0.1028556],"study_design_scores_gemma":[0.0003111777,0.0006187988,0.01221389,0.000009345769,0.000006118467,0.00001108556,0.00009163598,0.9279959,0.00003972551,0.00009041178,0.05839945,0.0002124618],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01342692,0.000017288,0.9794482,0.002559937,0.0003416867,0.00009515802,8.263303e-8,0.0002410844,0.003869673],"genre_scores_gemma":[0.9653999,0.000009713721,0.03051879,0.0001407145,0.0004555797,0.0000172272,0.00007971851,0.00001795861,0.003360393],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.951973,"threshold_uncertainty_score":0.7690423,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2150468603","doi":"10.1613/jair.3912","title":"The Arcade Learning Environment: An Evaluation Platform for General Agents","year":2013,"lang":"en","type":"article","venue":"Journal of Artificial Intelligence Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":1060,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"Alberta Innovates; University of Alberta; Compute Canada","keywords":"Testbed; Benchmarking; Reinforcement learning; Benchmark (surveying); Imitation; Interface (matter)","retraction":null,"screen_n_in":null,"score":{"opus":0.2902597031161736,"gpt":0.4427010381262976,"spread":0.152441335010124,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01036481,0.0001147353,0.0001458313,0.0002628453,0.0009449112,0.001006914,0.001638236,0.00007871621,0.0001490016],"category_scores_gemma":[0.001593545,0.00007997811,0.0001105602,0.0003240265,0.0001897001,0.001343404,0.0002195739,0.0008387021,0.0002262879],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002630366,"about_ca_system_score_gemma":0.0002240434,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002925329,"about_ca_topic_score_gemma":0.000003393742,"domain_scores_codex":[0.9958612,0.0004719288,0.0007582436,0.0002182003,0.002075787,0.0006146715],"domain_scores_gemma":[0.9968795,0.0009411229,0.0003592677,0.0003876486,0.001224912,0.000207604],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002452238,0.00003875276,0.00006165066,0.000004599262,0.0000231304,0.00000178245,0.001141798,0.5649804,0.002540559,0.01724453,0.0003115633,0.4136267],"study_design_scores_gemma":[0.00004503443,0.0008589922,0.0002236359,0.0000164548,0.00000575325,0.00001097539,0.0008975657,0.9479652,0.008820786,0.03571634,0.005351497,0.00008775616],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0776395,0.0001141506,0.9195682,0.001294114,0.0004011518,0.0006331351,1.521434e-7,0.00001266543,0.0003369687],"genre_scores_gemma":[0.9676027,0.0003382145,0.03057141,0.00004015376,0.0006171849,0.00006022278,0.000001990808,0.00001903796,0.0007490409],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8899633,"threshold_uncertainty_score":0.9709696,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2964227312","doi":"10.1609/aaai.v31i1.10916","title":"The Option-Critic Architecture","year":2017,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":695,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"Fonds de recherche du Québec – Nature et technologies; Natural Sciences and Engineering Research Council of Canada","keywords":"Reinforcement learning; Flexibility (engineering); Computer science; Abstraction; Architecture; Key (lock); Artificial intelligence; Machine learning; Computer security; Economics; Management; Epistemology","retraction":null,"screen_n_in":null,"score":{"opus":0.01411873004184306,"gpt":0.2683013584491692,"spread":0.2541826284073262,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.000167752,0.00005150459,0.00003813567,0.00001222153,0.001186485,0.001166134,0.001925212,0.00001841626,0.000009146883],"category_scores_gemma":[0.0001923645,0.00002995117,0.00002859321,0.00002088642,0.00008266725,0.0002116392,0.0003495716,0.0001062275,0.0001767067],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00000868247,"about_ca_system_score_gemma":0.00001906291,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001097338,"about_ca_topic_score_gemma":0.000005212677,"domain_scores_codex":[0.9994591,0.00001650614,0.00007638545,0.0001143495,0.0001672662,0.0001663662],"domain_scores_gemma":[0.9985939,0.00009728855,0.00005036699,0.001191545,0.00003160837,0.00003524238],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[6.5278e-7,0.000002478694,0.000277192,0.000002293115,0.00000679328,0.000003349486,0.0001084994,0.06727173,0.00005034822,0.8937619,0.00117932,0.03733539],"study_design_scores_gemma":[0.0001758966,0.00004970566,0.01018661,0.00001114816,0.000002736119,0.00001534647,0.00001655098,0.8550664,0.0003721338,0.01743314,0.1165144,0.0001559295],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0002532652,0.0000143036,0.93272,0.008261321,0.0003982317,0.00005038731,2.852388e-8,0.00008044368,0.058222],"genre_scores_gemma":[0.930691,0.00001430557,0.04795191,0.0003655719,0.00008126167,0.000004686381,1.752725e-7,0.000004130965,0.02088695],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9304377,"threshold_uncertainty_score":0.9998708,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2094387729","doi":"10.1016/j.automatica.2009.07.008","title":"Natural actor–critic algorithms","year":2009,"lang":"en","type":"article","venue":"Automatica","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":569,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Temporal difference learning; Reinforcement learning; Computer science; Function approximation; Convergence (economics); Bellman equation; Mathematical proof; Function (biology); Stochastic gradient descent; Variance (accounting); Mathematics; Mathematical optimization; Artificial intelligence; Applied mathematics; Algorithm; Artificial neural network","retraction":null,"screen_n_in":null,"score":{"opus":0.009214446772915929,"gpt":0.2557505181836024,"spread":0.2465360714106865,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000140243,0.0001259981,0.0001392311,0.00007468036,0.000102602,0.0002203867,0.0008319107,0.00004454845,0.00003426801],"category_scores_gemma":[0.0001332536,0.0001118786,0.00005637345,0.0002688565,0.00002844921,0.0004764191,0.00009320102,0.0001742394,0.0004491799],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004210856,"about_ca_system_score_gemma":0.00003856831,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000159756,"about_ca_topic_score_gemma":9.431805e-8,"domain_scores_codex":[0.9988358,0.00003611151,0.0002227011,0.0002165021,0.0003569981,0.000331858],"domain_scores_gemma":[0.9991421,0.0001027287,0.00005806637,0.0005655764,0.000048424,0.00008314654],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000004362511,0.0002060781,0.0002340662,0.0000800835,0.00009036154,0.0002083638,0.003729093,0.03878931,0.004153063,0.5334502,0.01907304,0.399982],"study_design_scores_gemma":[0.000156992,0.0001033798,0.007838191,0.00002352456,0.00000489476,0.00002317627,0.000008011419,0.9865984,0.0004123057,0.002861236,0.001803989,0.0001659013],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002371119,0.00006220524,0.9814356,0.003212282,0.0005603607,0.0001228167,1.51954e-7,0.0008934062,0.01134205],"genre_scores_gemma":[0.8310422,0.000002862118,0.1666463,0.00123374,0.00005489538,0.000002312721,0.000001650235,0.000005339909,0.001010732],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9478091,"threshold_uncertainty_score":0.5773448,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2626637010","doi":"10.4230/lipics.cp.2023.25","title":"Learning a Generic Value-Selection Heuristic Inside a Constraint Programming Solver","year":2017,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":523,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Decomposition; Value (mathematics); Computer science; Artificial intelligence; Machine learning; Chemistry","retraction":null,"screen_n_in":null,"score":{"opus":0.07875522306874225,"gpt":0.2092543842109096,"spread":0.1304991611421674,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004637241,0.0004410352,0.0004255934,0.0003504416,0.0007358423,0.0007127193,0.002079834,0.0003686937,0.00002062688],"category_scores_gemma":[0.0003057366,0.0005387974,0.0002673581,0.00036901,0.0002480129,0.0004863974,0.002449004,0.001527679,0.0001209387],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004356817,"about_ca_system_score_gemma":0.0004992771,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002202635,"about_ca_topic_score_gemma":0.00001682741,"domain_scores_codex":[0.9973414,0.0002438049,0.0003030628,0.001274081,0.0002072294,0.0006304086],"domain_scores_gemma":[0.9974022,0.0001240822,0.0007919993,0.001225121,0.000245076,0.0002115113],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000009386253,0.00002873752,0.003011122,0.00008415493,0.0001080708,0.0002534798,0.0002825991,0.9652758,0.00005054007,0.02693993,0.00004487991,0.003911263],"study_design_scores_gemma":[0.0003688268,0.0001491292,0.0008215543,0.0001630741,0.00009291099,0.0000248876,0.00005360494,0.9926903,0.00005123162,0.002324664,0.002688381,0.0005714926],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03408676,0.00003376661,0.9602556,0.00004906475,0.0007744258,0.0004189151,8.086069e-7,0.0006177943,0.003762871],"genre_scores_gemma":[0.9773498,0.00009790313,0.01883078,0.00004557729,0.0001277442,0.000002679999,0.00001460327,0.00003373277,0.003497223],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.943263,"threshold_uncertainty_score":0.9997063,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2144913588","doi":"10.1613/jair.2567","title":"Online Planning Algorithms for POMDPs","year":2008,"lang":"en","type":"article","venue":"Journal of Artificial Intelligence Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":515,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Université Laval; McGill University","funders":"National Institute of Mental Health; Fonds Québécois de la Recherche sur la Nature et les Technologies","keywords":"Partially observable Markov decision process; Computer science; Heuristic; Markov decision process; Mathematical optimization; Upper and lower bounds; Reduction (mathematics); Observable; Focus (optics); Computational complexity theory; Action (physics); Artificial intelligence; Machine learning; Markov process; Markov chain; Algorithm; Markov model; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.4539481331437369,"gpt":0.4870388551056944,"spread":0.0330907219619575,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003161627,0.0001193069,0.0002483997,0.0006251191,0.0004626185,0.0001813799,0.001643503,0.00008752154,0.00001869915],"category_scores_gemma":[0.001290521,0.0001021837,0.0001577771,0.0008386896,0.0001733774,0.0006108243,0.0002307723,0.0008163073,0.00004755243],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001424861,"about_ca_system_score_gemma":0.0005033809,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008837411,"about_ca_topic_score_gemma":8.872053e-7,"domain_scores_codex":[0.9968257,0.000170885,0.000840912,0.0002181855,0.001341275,0.0006030769],"domain_scores_gemma":[0.9963918,0.001072907,0.0002897877,0.0003483959,0.001682321,0.0002148163],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000167487,0.0007302961,0.0004819494,0.00005890309,0.0001337133,0.0006344534,0.008394283,0.7665037,0.003862735,0.0436602,0.004426053,0.1709463],"study_design_scores_gemma":[0.000070068,0.001535399,0.0001669656,0.00009343179,0.000004972575,0.0003390118,0.0007726636,0.9612532,0.01868797,0.00971067,0.007198069,0.0001675333],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01414195,0.0001975009,0.9834687,0.001156918,0.0005757028,0.00019062,9.831576e-7,0.00003347589,0.0002341798],"genre_scores_gemma":[0.5305747,0.0002793267,0.4673403,0.00009028836,0.001237464,0.000006051502,0.000001624137,0.00002336433,0.0004469132],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5164327,"threshold_uncertainty_score":0.4166929,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4224220194","doi":"10.1016/j.neunet.2022.03.037","title":"Deep learning, reinforcement learning, and world models","year":2022,"lang":"en","type":"review","venue":"Neural Networks","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":493,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Artificial intelligence; Computer science; Session (web analytics); Human intelligence; Deep learning; Cognitive science; Reinforcement; Psychology","retraction":null,"screen_n_in":null,"score":{"opus":0.04696882901226523,"gpt":0.2927112085831605,"spread":0.2457423795708953,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.0006544319,0.000849761,0.001538543,0.0004786976,0.0008028433,0.0006343161,0.001825254,0.0002887159,0.0004274015],"category_scores_gemma":[0.0000839915,0.0007896046,0.0004854276,0.001255128,0.00009776878,0.0005656304,0.002701472,0.004223017,0.00003438108],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002440002,"about_ca_system_score_gemma":0.00007803505,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001270443,"about_ca_topic_score_gemma":0.000003855707,"domain_scores_codex":[0.9951777,0.0007663104,0.001052129,0.00114287,0.000840757,0.001020177],"domain_scores_gemma":[0.9971245,0.0006886589,0.001030476,0.0008201958,0.00004532849,0.0002908426],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001082722,0.000003684602,0.000003175718,0.0004128048,0.00004322454,0.00002399019,0.00004199343,0.5494549,9.859455e-10,0.001698198,0.0002604373,0.4480565],"study_design_scores_gemma":[0.0000856254,0.0001317514,1.42694e-7,0.0003327031,0.00008425572,0.00003854027,0.000003243852,0.5022877,3.250746e-9,0.00001732846,0.4966387,0.0003799662],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"review","genre_gemma":"review","genre_scores_codex":[2.683847e-8,0.5497999,0.442696,0.00001436089,0.0006205231,0.000438554,7.871848e-8,0.000338992,0.006091516],"genre_scores_gemma":[0.0000861663,0.9745811,0.001226661,0.0001547552,0.0003219173,0.0001772599,0.0001121679,0.0001124396,0.02322754],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.4963783,"threshold_uncertainty_score":0.9994555,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2055921164","doi":"10.1007/s10458-012-9200-2","title":"A survey of point-based POMDP solvers","year":2012,"lang":"en","type":"article","venue":"Autonomous Agents and Multi-Agent Systems","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":433,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Partially observable Markov decision process; Markov decision process; Computer science; Bellman equation; Function (biology); Observable; Value (mathematics); Point (geometry); State space; Mathematical optimization; Theoretical computer science; Space (punctuation); Scale (ratio); Markov chain; Markov process; Mathematics; Markov model; Machine learning","retraction":null,"screen_n_in":null,"score":{"opus":0.06430470677447479,"gpt":0.2858437758454526,"spread":0.2215390690709778,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001216136,0.0002224049,0.0003408771,0.000140295,0.0001283623,0.0001267125,0.0005132077,0.00009352862,0.0000141041],"category_scores_gemma":[0.00007851667,0.0001979465,0.00007216677,0.0002090204,0.0000592844,0.0004050514,0.0002002165,0.0001114841,0.00006527548],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008877759,"about_ca_system_score_gemma":0.00008558315,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001867643,"about_ca_topic_score_gemma":0.00001281675,"domain_scores_codex":[0.9980686,0.00023435,0.0005365442,0.0003106401,0.0003368746,0.0005129657],"domain_scores_gemma":[0.9985338,0.0001364224,0.0003675697,0.0006006775,0.0001166007,0.0002449627],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001328268,0.000477245,0.7583708,0.0003878576,0.0002442389,0.00001337557,0.003726563,0.2262678,0.0006338895,0.002697127,0.003855247,0.003312567],"study_design_scores_gemma":[0.000767431,0.00007488515,0.2529084,0.00005041674,0.0000140189,0.000005158865,0.00006118571,0.741416,0.0001893309,2.886011e-7,0.00428455,0.0002283648],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05566288,0.000502663,0.9407591,0.00004546454,0.001855876,0.0005434519,0.0000127305,0.0001311061,0.0004867641],"genre_scores_gemma":[0.9934121,0.00001727051,0.005330982,0.0001310175,0.0000355932,0.00001808645,0.00001985863,0.00001945481,0.001015643],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9377492,"threshold_uncertainty_score":0.8072025,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2963923407","doi":"","title":"Addressing Function Approximation Error in Actor-Critic Methods","year":2018,"lang":"en","type":"article","venue":"UvA-DARE (University of Amsterdam)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":414,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Computer science; Bellman equation; Suite; Function (biology); Function approximation; Value (mathematics); Limit (mathematics); Artificial intelligence; Approximation algorithm; Temporal difference learning; Mathematical optimization; Machine learning; Algorithm; Artificial neural network; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.07700182874909071,"gpt":0.3215285431639732,"spread":0.2445267144148824,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004695347,0.0001167391,0.0001883031,0.0003085093,0.0001965967,0.000061242,0.0006376738,0.00009508116,0.0001140939],"category_scores_gemma":[0.00006597937,0.0001472616,0.00006337802,0.0005404582,0.0001573545,0.001170831,0.0003302252,0.0001625214,0.0000588998],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001143272,"about_ca_system_score_gemma":0.00005961961,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009073262,"about_ca_topic_score_gemma":0.00003730231,"domain_scores_codex":[0.998826,0.0002086075,0.0001579347,0.0003034289,0.0002640247,0.0002400642],"domain_scores_gemma":[0.9990621,0.00009876372,0.0001766799,0.0004361766,0.0001641826,0.00006210866],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004657088,0.0004408836,0.01589933,0.001021615,0.0002674565,0.000107197,0.06703836,0.06675688,0.04366093,0.05056656,0.004253956,0.7495211],"study_design_scores_gemma":[0.001074633,0.0004323008,0.04460939,0.0002488052,0.00003262333,0.000007439533,0.002492132,0.9436291,0.001198327,0.0009356977,0.004993854,0.0003456795],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03467278,0.00001569866,0.9586182,0.0003059115,0.0003963739,0.0001378391,7.969048e-7,0.0001037584,0.005748617],"genre_scores_gemma":[0.7351088,0.000002276131,0.2641078,0.00007124237,0.00003908871,1.837117e-7,0.000005780023,0.000006758804,0.0006580958],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8768722,"threshold_uncertainty_score":0.6005155,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2104641222","doi":"10.1177/105971230501300301","title":"Reinforcement Learning for RoboCup Soccer Keepaway","year":2005,"lang":"en","type":"article","venue":"Adaptive Behavior","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":390,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Generality; Computer science; Artificial intelligence; Learning classifier system; Benchmark (surveying); Machine learning; Task (project management); Psychology","retraction":null,"screen_n_in":null,"score":{"opus":0.03632614507992138,"gpt":0.2862749197792012,"spread":0.2499487746992798,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003240065,0.0002483064,0.0002247243,0.0001296592,0.0003251279,0.0001662337,0.0007878341,0.0001020431,0.0001092585],"category_scores_gemma":[0.00006378473,0.0002461059,0.0001664094,0.0002121845,0.0000566728,0.0007713095,0.0002897343,0.0002896718,0.000307009],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002095527,"about_ca_system_score_gemma":0.00008101299,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001499996,"about_ca_topic_score_gemma":0.000004185928,"domain_scores_codex":[0.9980931,0.00004767428,0.0003966616,0.0004679125,0.0004257464,0.0005689494],"domain_scores_gemma":[0.9988121,0.0001293928,0.0002117161,0.0005068859,0.0002002297,0.0001396327],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001602943,0.00004696429,0.0009961713,0.000005166805,0.0000240261,0.00000489849,0.0004977155,0.9213493,0.0006962757,0.03408218,0.0008843093,0.04139692],"study_design_scores_gemma":[0.001106256,0.0007734715,0.004317374,0.00002766958,0.00006573344,0.00001139385,0.00008802991,0.9103401,0.003698291,0.00005859257,0.07897341,0.000539645],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002160611,0.00004419385,0.9916484,0.0003521383,0.0002968804,0.0007363658,5.48967e-7,0.0003221587,0.004438723],"genre_scores_gemma":[0.8498898,0.000009882785,0.1321765,0.000367567,0.0001976642,0.000323979,0.00001267495,0.00003157423,0.01699037],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8594719,"threshold_uncertainty_score":0.9999991,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2132622533","doi":"10.5555/2031678.2031726","title":"Horde: a scalable real-time architecture for learning knowledge from unsupervised sensorimotor interaction","year":2011,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":305,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University; University of Alberta","funders":"","keywords":"Computer science; Artificial intelligence; Reinforcement learning; Function (biology); Scalability; Temporal difference learning; Modular design; Architecture; Robot; Robot learning; Robotics; Machine learning; Unsupervised learning; Mobile robot","retraction":null,"screen_n_in":null,"score":{"opus":0.03268646550292204,"gpt":0.2583288463899193,"spread":0.2256423808869972,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002363391,0.0002087182,0.000221136,0.000143473,0.0001930822,0.000153298,0.0006470305,0.0001090315,0.0004482751],"category_scores_gemma":[0.0001457549,0.000186533,0.0001141296,0.0002492749,0.00003124888,0.0004170176,0.0002340571,0.0002836924,0.0006485887],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000677533,"about_ca_system_score_gemma":0.00005828257,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002822767,"about_ca_topic_score_gemma":0.000009845169,"domain_scores_codex":[0.9985661,0.0001130016,0.0002922173,0.0004717264,0.0001735085,0.0003834594],"domain_scores_gemma":[0.9987479,0.0003474203,0.0001238735,0.000507862,0.0001525395,0.0001203571],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007093608,0.0007532553,0.008014528,0.0002994676,0.0007505495,0.00003762201,0.05778679,0.5234396,0.1183871,0.03280865,0.02131783,0.2356953],"study_design_scores_gemma":[0.0006285246,0.000380699,0.0009631351,0.00004712152,0.00001921573,0.000004833458,0.0001057292,0.9648424,0.007417363,0.0006224482,0.02464423,0.0003243015],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01438535,0.00001513361,0.9193181,0.00008789951,0.000447585,0.0003525553,0.000001177627,0.0005995876,0.0647926],"genre_scores_gemma":[0.2669595,0.00002739312,0.7031164,0.0001258418,0.0003045433,0.00006440374,0.00003903906,0.00005687371,0.02930601],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4414028,"threshold_uncertainty_score":0.8336512,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4353091694","doi":"10.1038/s41467-023-37180-x","title":"Catalyzing next-generation Artificial Intelligence through NeuroAI","year":2023,"lang":"en","type":"review","venue":"Nature Communications","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":283,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute; Ontario Brain Institute; Montreal Neurological Institute and Hospital","funders":"National Institute of Neurological Disorders and Stroke; National Institute of General Medical Sciences; National Eye Institute; National Institute of Mental Health; Natural Sciences and Engineering Research Council of Canada; Intelligence Advanced Research Projects Activity; Defense Advanced Research Projects Agency; Office of Naval Research; Cold Spring Harbor Laboratory; Multidisciplinary University Research Initiative; James S. McDonnell Foundation; Canadian Institute for Advanced Research; National Science Foundation; Semiconductor Research Corporation; National Institutes of Health; Lourie Foundation; Howard Hughes Medical Institute","keywords":"Computer science; Artificial intelligence; Data science","retraction":null,"screen_n_in":null,"score":{"opus":0.3747334748993592,"gpt":0.4284879568866756,"spread":0.05375448198731642,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","open_science","research_integrity","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0005107338,0.0003976944,0.0007144585,0.0002837183,0.0006787662,0.0007687197,0.008350063,0.0007243532,0.000005225339],"category_scores_gemma":[0.0007427941,0.0003781947,0.0003350507,0.002136254,0.000127263,0.0007766488,0.002792963,0.003109945,0.0009061987],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001723035,"about_ca_system_score_gemma":0.000395797,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001470587,"about_ca_topic_score_gemma":0.0000684054,"domain_scores_codex":[0.9972057,0.00049677,0.0009462154,0.0005722158,0.0004251,0.0003539828],"domain_scores_gemma":[0.9921865,0.001034754,0.0006131009,0.005900882,0.0001940795,0.00007061612],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[1.173269e-7,0.00002487569,1.089671e-7,0.0005358772,0.00006253833,0.000003023236,0.000225024,0.003668839,0.000001106732,0.3364345,0.001899124,0.6571449],"study_design_scores_gemma":[0.000007213488,0.00001515108,1.401097e-7,0.001037739,0.0001069564,0.00001310169,0.000009514318,0.1074357,0.000002375512,0.000542495,0.890515,0.000314637],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[1.529753e-8,0.6795169,0.3171865,0.000688215,0.000934313,0.0004512063,0.000005800267,0.0004182873,0.0007987869],"genre_scores_gemma":[0.00004266411,0.9429298,0.05543158,0.000187854,0.0002669623,0.0001393084,0.0006494219,0.00006073066,0.0002916915],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.8886158,"threshold_uncertainty_score":0.9998717,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2904453761","doi":"10.48550/arxiv.1812.02900","title":"Off-Policy Deep Reinforcement Learning without Exploration","year":2018,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":280,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Computer science; Extrapolation; Artificial intelligence; Reinforcement; Class (philosophy); Uncorrelated; Space (punctuation); Action (physics); Control (management); Machine learning; Mathematics; Engineering","retraction":null,"screen_n_in":null,"score":{"opus":0.0781399192839883,"gpt":0.2175502889724059,"spread":0.1394103696884176,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004376033,0.0004687819,0.0003984674,0.0005934703,0.0003966733,0.0003678288,0.00227604,0.0003665528,0.00006599739],"category_scores_gemma":[0.0001954335,0.0005605649,0.0002121429,0.0008241462,0.0001709339,0.001131763,0.003414909,0.000973916,0.0006518754],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006114824,"about_ca_system_score_gemma":0.0004254793,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001160187,"about_ca_topic_score_gemma":0.00001215629,"domain_scores_codex":[0.9973186,0.0002327417,0.000392993,0.001185167,0.0002704489,0.0006000365],"domain_scores_gemma":[0.9970809,0.00007720346,0.0006688882,0.001615139,0.0003282726,0.0002296291],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001702318,0.00001533724,0.001272256,0.00004695106,0.00007843606,0.00003759088,0.0007431455,0.9362316,0.00001148601,0.06067394,0.0001389214,0.0007333177],"study_design_scores_gemma":[0.000437006,0.0001884926,0.0001752539,0.0001105666,0.00004652889,0.000002832795,0.00009162052,0.9903445,0.000127813,0.004330835,0.003579814,0.0005647122],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009142648,0.00002239434,0.9742266,0.0001548998,0.0007084008,0.0004202038,3.367011e-7,0.000602242,0.01472229],"genre_scores_gemma":[0.9831737,0.0003554383,0.003483821,0.000160451,0.000360577,0.00000278623,0.0000442914,0.00003730545,0.01238158],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9740311,"threshold_uncertainty_score":0.9996846,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3107153805","doi":"10.1038/s41586-020-2939-8","title":"Autonomous navigation of stratospheric balloons using reinforcement learning","year":2020,"lang":"en","type":"article","venue":"Nature","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":272,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Google (Canada)","funders":"","keywords":"Reinforcement learning; Computer science; Controller (irrigation); Obstacle avoidance; Process (computing); Obstacle; Reinforcement; Real-time computing; Artificial intelligence; Simulation; Engineering","retraction":null,"screen_n_in":null,"score":{"opus":0.01814971253088397,"gpt":0.2613037861494292,"spread":0.2431540736185452,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001254197,0.0001233215,0.000153186,0.00002688758,0.00009266581,0.00007274236,0.0005520114,0.0003647256,0.00002667627],"category_scores_gemma":[0.0001020871,0.0001238732,0.00006203735,0.0005112364,0.00002586024,0.0003473492,0.0001903805,0.001480309,0.00001552104],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005458733,"about_ca_system_score_gemma":0.0001100498,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006870286,"about_ca_topic_score_gemma":1.158803e-7,"domain_scores_codex":[0.9988515,0.00004181936,0.0002708015,0.000238469,0.0003973989,0.0001999713],"domain_scores_gemma":[0.9993131,0.00003530781,0.0002346632,0.0002284794,0.0001085471,0.00007993104],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000003112049,0.000003845925,0.0005735282,0.00003470875,0.00001569003,0.00000398425,0.0006551655,0.9834239,0.004911462,0.009381237,0.00007731764,0.0009160832],"study_design_scores_gemma":[0.0001952881,0.0001455988,0.0003433974,0.00003555523,0.00000919907,0.000004364529,0.00005334196,0.9915751,0.004171271,0.0000325549,0.003305055,0.0001292555],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02558101,0.0001341505,0.9707657,0.0004715101,0.000194659,0.0001373047,2.227059e-7,0.0001662504,0.002549255],"genre_scores_gemma":[0.9481514,0.000003795101,0.05093965,0.0007097254,0.00005836394,0.000001240036,0.000006656787,0.000009954401,0.0001192131],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9225704,"threshold_uncertainty_score":0.6431289,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1556824961","doi":"10.1007/3-540-45622-8_16","title":"Learning Options in Reinforcement Learning","year":2002,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":270,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Computer science; Psychology; Artificial intelligence","retraction":null,"screen_n_in":null,"score":{"opus":0.02108976112327637,"gpt":0.2445734416842048,"spread":0.2234836805609284,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001448459,0.0006915959,0.0006557687,0.001862354,0.0004730661,0.0008107211,0.003371776,0.0004224241,0.0001405219],"category_scores_gemma":[0.0003523065,0.0007209074,0.0001421735,0.001159943,0.0004966671,0.0009313479,0.001998236,0.003139857,0.0002779543],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008368075,"about_ca_system_score_gemma":0.0003227981,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003021724,"about_ca_topic_score_gemma":0.00001797927,"domain_scores_codex":[0.994575,0.00009721199,0.0009717394,0.00166236,0.001544706,0.001148935],"domain_scores_gemma":[0.9972336,0.0005613707,0.0005350243,0.001248701,0.0002109309,0.0002104324],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001800288,0.000009474928,0.0001944999,0.00002449112,0.000007078985,0.00008053609,0.001000061,0.8959591,0.00002666361,0.01240253,0.000006909607,0.0902869],"study_design_scores_gemma":[0.0003435532,0.000415549,0.0001366579,0.0005290617,0.000005770522,0.00004033422,5.469884e-7,0.9873369,0.00006838211,0.004950698,0.005414336,0.0007581717],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00003850032,0.0001999942,0.974399,0.0003995299,0.001060396,0.0004346616,1.768159e-7,0.0003119211,0.02315585],"genre_scores_gemma":[0.6001106,0.0007126376,0.3726501,0.001051521,0.0006231231,0.00003717071,0.0000207032,0.000134026,0.0246601],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6017488,"threshold_uncertainty_score":0.9995242,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1600046456","doi":"","title":"Off-Policy Temporal Difference Learning with Function Approximation","year":2001,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":254,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Temporal difference learning; Computer science; Function approximation; Function (biology); Artificial intelligence; Reinforcement learning; Artificial neural network","retraction":null,"screen_n_in":null,"score":{"opus":0.01561973101356947,"gpt":0.2314765468665761,"spread":0.2158568158530067,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001513505,0.00012133,0.00009936544,0.0001461771,0.0001631823,0.0002100491,0.0003344572,0.00004377302,0.00003732549],"category_scores_gemma":[0.000051394,0.00009231205,0.00002308696,0.0005853257,0.00003022502,0.0005203834,0.0001031207,0.0001846576,0.0001131415],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005544786,"about_ca_system_score_gemma":0.00006954476,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006489439,"about_ca_topic_score_gemma":0.000006579305,"domain_scores_codex":[0.9989574,0.00005043814,0.0001620384,0.0002522212,0.0003323207,0.0002455905],"domain_scores_gemma":[0.9993996,0.00003939699,0.0001120166,0.0003083909,0.00007594762,0.0000646746],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002740299,0.0000318824,0.07366014,0.0000138534,0.00002018632,0.000006965558,0.000439335,0.700115,0.0003113858,0.08705508,0.00009836206,0.1382204],"study_design_scores_gemma":[0.0002902564,0.0003406899,0.02185859,0.00001466385,0.00000377003,0.00002140267,0.00004364315,0.9702744,0.00008383077,0.0003519406,0.006556002,0.0001607357],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01558563,0.00000786569,0.9553232,0.0004910995,0.00007130956,0.0001127137,1.922036e-8,0.0004302774,0.02797789],"genre_scores_gemma":[0.9432571,0.00001317453,0.03908758,0.0002219266,0.00008110137,0.000009377155,0.000005918902,0.000008985148,0.01731485],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9276714,"threshold_uncertainty_score":0.3764376,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1496855202","doi":"10.48550/arxiv.1301.6690","title":"Model-Based Bayesian Exploration","year":2013,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":235,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Reinforcement learning; Value of information; Action (physics); Bayesian probability; Quality (philosophy); Artificial intelligence; Value (mathematics); Machine learning; Probability distribution; Mathematical optimization; Mathematics; Statistics","retraction":null,"screen_n_in":null,"score":{"opus":0.07631130262408777,"gpt":0.1767027802933835,"spread":0.1003914776692957,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00007955314,0.0001103527,0.00008587811,0.0001238392,0.0001268953,0.0001180858,0.0006973111,0.00005624745,0.00005628883],"category_scores_gemma":[0.00001787389,0.0001238894,0.00005378835,0.0004509061,0.0000401428,0.001623452,0.0001225139,0.000106848,0.000526075],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007370902,"about_ca_system_score_gemma":0.00006411895,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002645691,"about_ca_topic_score_gemma":0.000003004254,"domain_scores_codex":[0.9992237,0.00004311014,0.00009923644,0.0003365224,0.0000725649,0.0002248799],"domain_scores_gemma":[0.9991118,0.00003861427,0.00007684858,0.000562534,0.0001027311,0.0001074643],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001426245,0.00001136145,0.0005378191,0.000003535274,0.000004370388,0.000007356103,0.00005994866,0.8980603,0.00007524983,0.1006505,0.0003345115,0.0002536747],"study_design_scores_gemma":[0.0002475162,0.00004017923,0.0001588941,0.000006533234,0.000005108656,3.237427e-7,0.00002290058,0.992088,0.000187328,0.006970929,0.0001245454,0.0001477634],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01103663,0.000001714414,0.9833527,0.0002670486,0.00009162931,0.0001465984,1.595947e-7,0.0002309186,0.004872598],"genre_scores_gemma":[0.9812556,0.000003924738,0.01582843,0.0002552535,0.00001426493,8.150006e-7,0.000002776367,0.000007546967,0.002631409],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.970219,"threshold_uncertainty_score":0.6761805,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2236244207","doi":"10.1561/2200000049","title":"Bayesian Reinforcement Learning: A Survey","year":2015,"lang":"en","type":"article","venue":"Foundations and Trends® in Machine Learning","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":223,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Machine learning; Reinforcement learning; Artificial intelligence; Bayesian inference; Bayesian probability; Variable-order Bayesian network; Prior probability; Inference; Bellman equation; Algorithm; Mathematical optimization; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.0465390705497981,"gpt":0.3022379049415592,"spread":0.2556988343917611,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001479584,0.0002161315,0.0002320648,0.0006192418,0.0003316178,0.0004510015,0.0004007877,0.00007362461,0.0001070848],"category_scores_gemma":[0.0006175301,0.000215479,0.00004370805,0.001155608,0.00005793959,0.0006242121,0.0003474737,0.0006992697,0.00006496777],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001080186,"about_ca_system_score_gemma":0.00006361707,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00113169,"about_ca_topic_score_gemma":0.0003716036,"domain_scores_codex":[0.9979447,0.0004275327,0.0004201437,0.0004207795,0.0003861769,0.0004006943],"domain_scores_gemma":[0.9989625,0.0002076822,0.0001983828,0.0003276474,0.0001085285,0.0001952816],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000007608463,0.00001615743,0.1566369,0.000004218743,0.00001376265,0.000007336901,0.001010488,0.8160558,0.000002830401,0.005321153,0.00009279529,0.02083096],"study_design_scores_gemma":[0.000790716,0.000207591,0.04959874,0.00001674413,0.000005486397,0.00001307129,0.00006400618,0.9257292,0.000002412519,0.0001554297,0.02317936,0.0002372948],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00440627,0.0001102843,0.9754209,0.0005987748,0.0002232921,0.00009258708,4.754638e-7,0.0002337435,0.0189137],"genre_scores_gemma":[0.9800444,0.00002885283,0.008560605,0.00006780464,0.00003635856,0.00001436065,0.0002063629,0.00001953804,0.01102169],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9756382,"threshold_uncertainty_score":0.8786981,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2523728418","doi":"10.1609/aaai.v31i1.10916","title":"The Option-Critic Architecture","year":2017,"lang":"en","type":"preprint","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":209,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"Fonds de recherche du Québec – Nature et technologies; Natural Sciences and Engineering Research Council of Canada","keywords":"Reinforcement learning; Flexibility (engineering); Computer science; Architecture; Abstraction; Key (lock); Artificial intelligence; Economics; Computer security; Management; Epistemology","retraction":null,"screen_n_in":null,"score":{"opus":0.08507794004948784,"gpt":0.3192055271439572,"spread":0.2341275870944694,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","open_science"],"consensus_categories":[],"category_scores_codex":[0.001018551,0.0004552761,0.0004140548,0.0001252389,0.001155864,0.002453403,0.01123567,0.0002800105,0.00001785907],"category_scores_gemma":[0.001889532,0.0002841477,0.0003085877,0.0002131583,0.000880312,0.0002583385,0.004104813,0.001772007,0.0001240684],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008922501,"about_ca_system_score_gemma":0.0003429487,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000462813,"about_ca_topic_score_gemma":0.00000970352,"domain_scores_codex":[0.9967235,0.00003970655,0.0008099793,0.0007982968,0.001054284,0.0005742543],"domain_scores_gemma":[0.9955984,0.0002900421,0.001316115,0.001578888,0.001104034,0.0001125183],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00002288254,0.00003127514,0.00002810786,0.0001048706,0.00003791715,6.255983e-7,0.0009051112,0.0308022,0.0009967545,0.9113905,0.0002319966,0.05544769],"study_design_scores_gemma":[0.00001758959,0.0001182962,0.00009237354,0.0008913585,0.0000289177,0.000005808616,0.0001584817,0.3900288,0.04096578,0.566276,0.001025061,0.0003915443],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007303503,0.0001725897,0.9086833,0.03307563,0.005435375,0.001698945,0.000009606189,0.0003039365,0.04331713],"genre_scores_gemma":[0.992047,0.000248137,0.005288757,0.000167745,0.0002310782,0.00007133481,0.000001053967,0.00002564628,0.001919317],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9847434,"threshold_uncertainty_score":0.9999611,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2174786457","doi":"10.48550/arxiv.1511.06342","title":"Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning","year":2015,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":208,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"","keywords":"Reinforcement learning; Computer science; Exploit; Artificial intelligence; Transfer of learning; Set (abstract data type); Task (project management); Machine learning; Engineering","retraction":null,"screen_n_in":null,"score":{"opus":0.07661995159032055,"gpt":0.1959416640452457,"spread":0.1193217124549251,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004393147,0.0004550408,0.0004196315,0.0003037256,0.0002267781,0.0002756433,0.001556099,0.0003733191,0.00003523204],"category_scores_gemma":[0.00008195826,0.0005224983,0.0001596511,0.0003528064,0.0001470049,0.000578645,0.002137219,0.001185823,0.0001062439],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003332755,"about_ca_system_score_gemma":0.0002078969,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009791936,"about_ca_topic_score_gemma":0.00001593515,"domain_scores_codex":[0.9976581,0.0001763913,0.0003109058,0.001103297,0.0002206087,0.0005307141],"domain_scores_gemma":[0.9980758,0.0001141297,0.000180699,0.001043803,0.0002419605,0.0003436219],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001989643,0.00001524419,0.001644316,0.0000749872,0.00009771503,0.00009938002,0.0008383752,0.9728162,0.00002740382,0.02389677,0.00007999845,0.0003897577],"study_design_scores_gemma":[0.0007869599,0.0001416586,0.000284112,0.00008431375,0.00007446969,0.000005761398,0.0001428235,0.9946031,0.00006847557,0.001025683,0.002199968,0.0005826941],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05056662,0.0000724946,0.9436437,0.00005909443,0.0005010002,0.0003599242,0.000001085686,0.0003674588,0.004428579],"genre_scores_gemma":[0.9930797,0.0003131168,0.002293247,0.00008640963,0.00006315284,0.000001270768,0.00002706163,0.00003075128,0.00410533],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.942513,"threshold_uncertainty_score":0.9997227,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W190427941","doi":"10.20965/jaciii.2006.p0578","title":"Opposition-Based Reinforcement Learning","year":2006,"lang":"en","type":"article","venue":"Journal of Advanced Computational Intelligence and Intelligent Informatics","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":207,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Waterloo","funders":"","keywords":"Reinforcement learning; Computer science; Expediting; Artificial intelligence; A priori and a posteriori; Probabilistic logic; Grid; Machine learning; Opposition (politics); Convergence (economics); Learning classifier system; Engineering; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.01614136464924268,"gpt":0.2628008458279594,"spread":0.2466594811787167,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005498489,0.0002306714,0.0002937219,0.0004271247,0.0002118487,0.0002999356,0.0006300904,0.00007062213,0.00002680802],"category_scores_gemma":[0.0001218115,0.0002071642,0.0001410916,0.0004091804,0.00009627332,0.001508154,0.0001231656,0.0004298322,0.00003695668],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001333639,"about_ca_system_score_gemma":0.0001893375,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003933322,"about_ca_topic_score_gemma":3.057756e-7,"domain_scores_codex":[0.9970205,0.00004714569,0.001684827,0.0001173148,0.0008316338,0.0002985694],"domain_scores_gemma":[0.9971184,0.0004653713,0.001237984,0.0001822251,0.0008631263,0.0001329653],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000166519,0.00003243827,0.0001371404,0.00004180925,0.00002345207,0.000006417903,0.0004112443,0.9284414,0.00002472308,0.05218075,0.0001128595,0.01857112],"study_design_scores_gemma":[0.0001901074,0.0006821066,0.0000877072,0.0001478156,0.00001389779,0.00009786877,0.000423391,0.9779323,0.003763405,0.01207858,0.004354897,0.0002279639],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002484691,0.0001775767,0.9947917,0.0002583821,0.0003203919,0.0001457123,3.951245e-7,0.00005071062,0.00177046],"genre_scores_gemma":[0.712014,0.000119827,0.2872182,0.0004158129,0.00006795237,0.000002784506,0.0000132802,0.000009202067,0.0001388741],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7095293,"threshold_uncertainty_score":0.8447911,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3175558129","doi":"10.1609/aaai.v35i12.17276","title":"Improving Sample Efficiency in Model-Free Reinforcement Learning from Images","year":2021,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":201,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Robustness (evolution); Computer science; Artificial intelligence; Stability (learning theory); Machine learning; Encoder; Representation (politics); Noise (video); Code (set theory); Image (mathematics)","retraction":null,"screen_n_in":null,"score":{"opus":0.01918541106757507,"gpt":0.2364670789210525,"spread":0.2172816678534774,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003401872,0.0001889616,0.0002121069,0.0001290999,0.0001490738,0.0003498227,0.001193542,0.00007138411,0.0001303387],"category_scores_gemma":[0.001040022,0.0001903449,0.00007206883,0.0005173096,0.00003073042,0.0006324667,0.001483019,0.0003685573,0.00004491041],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001265539,"about_ca_system_score_gemma":0.0002101923,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007857587,"about_ca_topic_score_gemma":0.00003856934,"domain_scores_codex":[0.9979848,0.0000757864,0.0004448449,0.0005542304,0.0004433271,0.0004969796],"domain_scores_gemma":[0.9982879,0.0003288337,0.0001328105,0.001045248,0.0001121638,0.00009301617],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001690143,0.00001546216,0.002173952,0.000008029428,0.000005836173,0.00001594585,0.0003893141,0.9806615,0.002360051,0.01152122,0.0001284205,0.002718571],"study_design_scores_gemma":[0.0005247236,0.00004483081,0.0005192518,0.00002398881,0.000003737429,0.000001846428,0.00007204143,0.9921091,0.004825045,0.001455438,0.0001816641,0.0002383008],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002079017,0.00008132191,0.9859313,0.0004868007,0.0001821406,0.0001074023,6.640364e-7,0.0002480983,0.01088325],"genre_scores_gemma":[0.6202108,0.00002670959,0.3755634,0.0003490996,0.00003364306,0.00001006844,0.00001966113,0.00001538164,0.003771283],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6181317,"threshold_uncertainty_score":0.7762042,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2102847492","doi":"10.1145/1390156.1390286","title":"Apprenticeship learning using linear programming","year":2008,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":194,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Apprenticeship; Markov decision process; Computer science; Linear programming; Solver; Frame (networking); Markov process; Mathematical optimization; Process (computing); Markov chain; Artificial intelligence; Machine learning; Algorithm; Mathematics; Statistics","retraction":null,"screen_n_in":null,"score":{"opus":0.06668564856124883,"gpt":0.2809145654987666,"spread":0.2142289169375178,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002190368,0.000120148,0.0001174947,0.00009438318,0.00036778,0.0001091385,0.0005981284,0.00005029282,0.00003030289],"category_scores_gemma":[0.0001003205,0.0001122962,0.00005762203,0.0004000482,0.00005220264,0.0005436887,0.0002881753,0.0002565385,0.000129726],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003284488,"about_ca_system_score_gemma":0.00005401459,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002577154,"about_ca_topic_score_gemma":4.463867e-7,"domain_scores_codex":[0.9987568,0.00005713522,0.0002136119,0.0002730321,0.0003440049,0.0003553867],"domain_scores_gemma":[0.999343,0.00006244618,0.0001051711,0.0003339181,0.00006871622,0.0000867997],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001128855,0.00001479226,0.01391528,0.000011243,0.00001143352,0.00002560094,0.0007192069,0.9762692,0.0002105855,0.003872871,0.00004250978,0.004906121],"study_design_scores_gemma":[0.0001420662,0.0000529735,0.0004208586,0.00001149192,0.000003455639,0.00006403509,0.00003248982,0.9878993,0.0004858464,0.00001288818,0.01071963,0.0001549848],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02527485,0.00002128833,0.9683531,0.00006256111,0.0001892532,0.0001043227,1.036349e-8,0.0005239496,0.00547068],"genre_scores_gemma":[0.5470853,0.00000504924,0.4498993,0.00007084916,0.00006273286,0.000002121358,8.50477e-7,0.000009465325,0.002864302],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5218105,"threshold_uncertainty_score":0.4579307,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1998172110","doi":"10.1007/s10479-005-5732-z","title":"Basis Function Adaptation in Temporal Difference Reinforcement Learning","year":2005,"lang":"en","type":"article","venue":"Annals of Operations Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":190,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Markov decision process; Bellman equation; Temporal difference learning; Basis function; Function approximation; Mathematical optimization; Basis (linear algebra); Computer science; Theory of computation; Q-learning; Convergence (economics); Markov process; Mathematics; Artificial intelligence; Algorithm; Artificial neural network","retraction":null,"screen_n_in":null,"score":{"opus":0.2255396635006515,"gpt":0.4118928531180281,"spread":0.1863531896173766,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001795468,0.00009287867,0.0001281759,0.0005990842,0.0002725683,0.0002002976,0.0005048928,0.00005984934,0.00008719236],"category_scores_gemma":[0.0004803068,0.00009155396,0.00003942791,0.001015105,0.00006737973,0.0008712418,0.000208319,0.0004367151,0.0001047277],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005640941,"about_ca_system_score_gemma":0.0001910033,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004123539,"about_ca_topic_score_gemma":0.000230164,"domain_scores_codex":[0.9977658,0.000321806,0.0004344844,0.0002619718,0.0008473764,0.0003685946],"domain_scores_gemma":[0.9986761,0.0001499368,0.00004510645,0.0003815484,0.000675984,0.00007130287],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001065656,0.00003482899,0.00141161,0.000008560532,0.000007176196,7.922054e-7,0.001046489,0.9596584,0.0006477443,0.0179498,0.0002057346,0.01901819],"study_design_scores_gemma":[0.0001889004,0.0003406269,0.007907618,0.00003693445,8.063308e-7,6.671843e-7,0.000214714,0.9867536,0.002050462,0.00006248661,0.002354988,0.00008817683],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05995004,0.00007519144,0.9328964,0.003314867,0.00005075815,0.0002850705,2.269717e-7,0.00004330525,0.003384158],"genre_scores_gemma":[0.9845845,0.0001614608,0.01089541,0.0001073689,0.00004718338,0.00004593613,0.00001631987,0.000007037598,0.004134838],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9246344,"threshold_uncertainty_score":0.3733462,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3117215073","doi":"10.1613/jair.1.13673","title":"Towards Continual Reinforcement Learning: A Review and Perspectives","year":2022,"lang":"en","type":"review","venue":"Journal of Artificial Intelligence Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":180,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal; McGill University","funders":"Canada Excellence Research Chairs, Government of Canada; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Computer science; Scope (computer science); Artificial intelligence; Bridging (networking); Taxonomy (biology); Function (biology); Management science; Engineering","retraction":null,"screen_n_in":null,"score":{"opus":0.3537488428002021,"gpt":0.4758490705427069,"spread":0.1221002277425048,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.01167099,0.0004019499,0.001920793,0.001212323,0.0005446997,0.0005758179,0.002878048,0.000173706,0.0008247282],"category_scores_gemma":[0.004848099,0.0003204127,0.000599543,0.00201187,0.0003637585,0.0005506512,0.001777283,0.004585868,0.0000898108],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000670457,"about_ca_system_score_gemma":0.001834283,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001694632,"about_ca_topic_score_gemma":7.451476e-7,"domain_scores_codex":[0.991866,0.002030564,0.002180656,0.0005497446,0.002603675,0.0007693485],"domain_scores_gemma":[0.994913,0.001465323,0.001499777,0.0006523895,0.001139064,0.0003304786],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000006593099,0.00004404091,1.560825e-7,0.006680058,0.0001705447,0.0001121582,0.001133247,0.002921728,2.790719e-7,0.02010213,0.000697781,0.9681313],"study_design_scores_gemma":[0.00001913925,0.001293918,1.013555e-7,0.008857085,0.0002132788,0.0003924475,0.00108381,0.003642684,0.000004018696,0.0005308807,0.983661,0.0003016187],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[1.006542e-7,0.7782069,0.21892,0.0005247617,0.0002594229,0.0006263192,5.310488e-7,0.00002235593,0.00143958],"genre_scores_gemma":[0.00002268204,0.9961339,0.00233357,0.00003572108,0.0002699145,0.00003875874,0.000002744795,0.00003634514,0.001126314],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9829633,"threshold_uncertainty_score":0.9999248,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2097113539","doi":"10.1613/jair.898","title":"Accelerating Reinforcement Learning through Implicit Imitation","year":2003,"lang":"en","type":"article","venue":"Journal of Artificial Intelligence Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":176,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto; University of British Columbia","funders":"Instituto de Ciencias del Mar y Limnología, Universidad Nacional Autónoma de México; Natural Sciences and Engineering Research Council of Canada","keywords":"Imitation; Reinforcement learning; Observability; Computer science; Convergence (economics); Action (physics); Space (punctuation); Artificial intelligence; State space; Reinforcement; Value (mathematics); Human–computer interaction; Machine learning; Psychology; Mathematics; Social psychology","retraction":null,"screen_n_in":null,"score":{"opus":0.2541885736812509,"gpt":0.4324161648659641,"spread":0.1782275911847133,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00631117,0.0001536442,0.0002507791,0.0004431295,0.0005890947,0.0007183951,0.001111902,0.000095398,0.0001312823],"category_scores_gemma":[0.00374955,0.0001367928,0.0001367768,0.001279598,0.0001224546,0.001501859,0.0001908028,0.001396905,0.0001742293],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002532483,"about_ca_system_score_gemma":0.0004110012,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002620821,"about_ca_topic_score_gemma":0.00000234679,"domain_scores_codex":[0.9954541,0.0006783106,0.001170682,0.0002570804,0.001750634,0.0006891928],"domain_scores_gemma":[0.9964403,0.0009106894,0.0005219298,0.0003631251,0.001603974,0.0001599956],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001358498,0.00002792562,0.00009031263,0.00001211117,0.00002187909,0.00002348281,0.002549058,0.7210429,0.006099456,0.2478757,0.000119119,0.02212445],"study_design_scores_gemma":[0.00009995599,0.001855639,0.0000452787,0.0001654187,0.000009800402,0.0001402498,0.006143613,0.7771646,0.1509272,0.05645241,0.006663742,0.0003321527],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007117496,0.0001014125,0.9806769,0.0005295918,0.0004407492,0.0001755816,4.165879e-8,0.00002476661,0.0109334],"genre_scores_gemma":[0.9279053,0.0001594595,0.0712276,0.00005820607,0.0001818653,0.000004634013,5.258775e-7,0.00001494361,0.0004474337],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9207879,"threshold_uncertainty_score":0.69275,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W13294968","doi":"10.1016/0021-8707(56)90033-8","title":"Toward Off-Policy Learning Control with Function Approximation","year":2010,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":175,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Temporal difference learning; Function approximation; Approximation algorithm; Function (biology); Computer science; Linear approximation; Bellman equation; Optimal control; Extension (predicate logic); Mathematical optimization; Control (management); Artificial intelligence; Mathematics; Reinforcement learning; Artificial neural network; Nonlinear system","retraction":null,"screen_n_in":null,"score":{"opus":0.02266731972468504,"gpt":0.2706007747695721,"spread":0.247933455044887,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005058447,0.0002650122,0.0002032879,0.0004177853,0.0002742498,0.0006056678,0.0009147468,0.0001129475,0.0003406384],"category_scores_gemma":[0.0006329065,0.0002305235,0.00006914238,0.000303596,0.0000715315,0.0007366529,0.000141652,0.001760627,0.0002702002],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000818189,"about_ca_system_score_gemma":0.0001684062,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008420838,"about_ca_topic_score_gemma":0.00001520498,"domain_scores_codex":[0.9978631,0.0001563385,0.0003251315,0.000490737,0.0008307642,0.0003339077],"domain_scores_gemma":[0.9986272,0.000168478,0.0003697274,0.0003038158,0.0004149482,0.0001158616],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001057528,0.00003190959,0.009701381,0.000008544373,0.00006762356,0.00000943245,0.0003967693,0.5157424,0.003923931,0.4264555,0.00001624982,0.04354053],"study_design_scores_gemma":[0.001065407,0.0005315858,0.003423867,0.00004001716,0.000009685872,0.00003256329,0.00004897326,0.9786261,0.0001885119,0.0009824905,0.01478053,0.0002702108],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0106375,0.000006076412,0.9326913,0.004855052,0.0006330823,0.0001860045,0.000001122295,0.0004629931,0.05052684],"genre_scores_gemma":[0.9877251,0.00001586709,0.007810126,0.000462858,0.0003230211,0.00002878056,0.00005251193,0.00002618901,0.003555513],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9770876,"threshold_uncertainty_score":0.9400476,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3092156990","doi":"10.1613/jair.1.12440","title":"Reward Machines: Exploiting Reward Function Structure in Reinforcement Learning","year":2022,"lang":"en","type":"article","venue":"Journal of Artificial Intelligence Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":172,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Vector Institute; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; University of Toronto; Government of Canada; Canadian Institute for Advanced Research; Vector Institute; Microsoft Research","keywords":"Reinforcement learning; Computer science; Exploit; Counterfactual thinking; Function (biology); Artificial intelligence; Finite-state machine; Machine learning; Algorithm; Psychology","retraction":null,"screen_n_in":null,"score":{"opus":0.1113564636583909,"gpt":0.3745631904237698,"spread":0.2632067267653789,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.007680622,0.000173457,0.0003079037,0.001429012,0.000805173,0.0003705154,0.00171371,0.00007099303,0.0004823425],"category_scores_gemma":[0.001497566,0.0001667666,0.0001175067,0.002283619,0.00009991034,0.0008412846,0.001028001,0.003419774,0.00003401743],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007225922,"about_ca_system_score_gemma":0.0004233338,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008558952,"about_ca_topic_score_gemma":0.00001439829,"domain_scores_codex":[0.9937354,0.00110426,0.001332225,0.0003433353,0.002735107,0.0007496561],"domain_scores_gemma":[0.9975918,0.0005915556,0.0005737434,0.0004105511,0.0006643192,0.000168032],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001468123,0.00003570172,0.0008341765,0.0000192612,0.00002008649,0.0001063135,0.002774453,0.9203317,0.003979521,0.01172592,0.0001459679,0.05988015],"study_design_scores_gemma":[0.0001137608,0.002815373,0.0001975151,0.0001026192,0.000006892068,0.0001379363,0.007290337,0.9629144,0.004042804,0.01653883,0.00557523,0.0002643098],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06806064,0.0001578907,0.9287345,0.001028922,0.001010319,0.0002438727,3.459332e-7,0.00003746507,0.0007260342],"genre_scores_gemma":[0.9942685,0.00007585859,0.004872578,0.00007221857,0.0002984882,0.00001080453,0.000002357972,0.00002001484,0.0003791165],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9262079,"threshold_uncertainty_score":0.9988794,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2134042548","doi":"","title":"Convergent Temporal-Difference Learning with Arbitrary Smooth Function Approximation","year":2009,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":168,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta; McGill University","funders":"","keywords":"Temporal difference learning; Function approximation; Stochastic approximation; Markov decision process; Mathematics; Function (biology); Stochastic gradient descent; Bellman equation; Mathematical optimization; Convergence (economics); Nonlinear system; Approximation algorithm; Artificial neural network; Gradient descent; Approximation error; Markov process; Rate of convergence; Algorithm; Applied mathematics; Reinforcement learning; Computer science; Artificial intelligence; Key (lock)","retraction":null,"screen_n_in":null,"score":{"opus":0.01454759173183138,"gpt":0.2203356894460717,"spread":0.2057880977142404,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003086166,0.0002223608,0.0002009777,0.0002202125,0.0003804546,0.001254692,0.000394027,0.00009119268,0.000002955897],"category_scores_gemma":[0.00003746253,0.0001781739,0.0000342085,0.0005714304,0.00002908029,0.005834027,0.00003667051,0.0003370327,0.00005803531],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008591708,"about_ca_system_score_gemma":0.0001059508,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001664588,"about_ca_topic_score_gemma":2.602174e-7,"domain_scores_codex":[0.9981189,0.00007744353,0.0006156633,0.0002124372,0.0006704886,0.0003050246],"domain_scores_gemma":[0.9985921,0.00002447851,0.0007143242,0.0002707612,0.0003032678,0.00009501889],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003811504,0.00002140727,0.002064663,0.0003278746,0.00001071911,0.000001978687,0.002588566,0.8960687,0.0002584095,0.007980965,0.000198037,0.09044057],"study_design_scores_gemma":[0.000384239,0.0004389012,0.004044361,0.0001566986,0.000007907748,0.0000324485,0.0002760931,0.9909536,0.0001483234,0.00005426573,0.003259605,0.0002435369],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01403097,0.000066004,0.9802634,0.0002932832,0.0003826908,0.0003897695,2.952454e-7,0.0007391103,0.003834518],"genre_scores_gemma":[0.9948147,0.000003605885,0.004097127,0.0004879033,0.0000631717,0.00002462385,0.00005705701,0.000007337842,0.0004444992],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9807837,"threshold_uncertainty_score":0.9997821,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2964295739","doi":"","title":"Imagination-Augmented Agents for Deep Reinforcement Learning","year":2017,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":163,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"","keywords":"Reinforcement learning; Robustness (evolution); Computer science; Artificial intelligence; Construct (python library); Policy learning; Machine learning; Contrast (vision); Deep learning; Context (archaeology); Architecture","retraction":null,"screen_n_in":null,"score":{"opus":0.0722518834787979,"gpt":0.2207498508139353,"spread":0.1484979673351374,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002569809,0.0001626348,0.0001428837,0.0001286689,0.001086613,0.0003607844,0.001712076,0.0000586294,0.00004311657],"category_scores_gemma":[0.0002108184,0.000190562,0.0001080869,0.0001424806,0.00008996658,0.001305494,0.0005407777,0.0001577804,0.000121041],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001535904,"about_ca_system_score_gemma":0.00004477881,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002914325,"about_ca_topic_score_gemma":0.000004092922,"domain_scores_codex":[0.9988506,0.00003764438,0.0001593805,0.0004698528,0.0001155496,0.0003669108],"domain_scores_gemma":[0.9983019,0.00007673725,0.0003472595,0.0009628466,0.0001930266,0.0001182889],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001083843,0.00001176179,0.004639879,0.00001196691,0.00003420143,0.00001949459,0.00011821,0.8581836,0.00003057792,0.1360206,0.0001987558,0.0007201487],"study_design_scores_gemma":[0.001031455,0.0001106828,0.002542647,0.00001765658,0.00002170997,0.000001309802,0.00004501782,0.9879441,0.0001831261,0.001090755,0.006795222,0.0002163043],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006112494,0.000005699126,0.9831771,0.0001910183,0.0003516741,0.0002896061,2.392711e-7,0.0001624218,0.009709782],"genre_scores_gemma":[0.9774202,0.00002840825,0.003015302,0.0001024224,0.00003797056,0.000001696441,0.000009398245,0.00001239997,0.01937227],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9801618,"threshold_uncertainty_score":0.8357456,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2011233848","doi":"10.1145/1143844.1143901","title":"Automatic basis function construction for approximate dynamic programming and reinforcement learning","year":2006,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":162,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Bellman equation; Markov decision process; Reinforcement learning; Computer science; Basis (linear algebra); Function approximation; Dynamic programming; Curse of dimensionality; Basis function; Mathematical optimization; State space; Temporal difference learning; Markov process; Dimensionality reduction; Q-learning; Function (biology); Space (punctuation); Artificial intelligence; Algorithm; Mathematics; Artificial neural network","retraction":null,"screen_n_in":null,"score":{"opus":0.007772714293839559,"gpt":0.2228768719720251,"spread":0.2151041576781855,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003167422,0.0001462463,0.0001386325,0.0001304069,0.0003020079,0.0003277692,0.0001577913,0.00005818046,0.00001730117],"category_scores_gemma":[0.00003346111,0.0001375079,0.0000471305,0.0002063135,0.00005133558,0.0004999363,0.0001014569,0.0001054299,0.00000830102],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007394009,"about_ca_system_score_gemma":0.00002413159,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002481319,"about_ca_topic_score_gemma":0.000002966593,"domain_scores_codex":[0.9988219,0.00002864943,0.0003235354,0.0002978648,0.0002204752,0.0003076362],"domain_scores_gemma":[0.9994066,0.00008116142,0.0001813806,0.0002097765,0.00007789013,0.00004314812],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000006091732,0.00001052799,0.001597743,0.0001376116,0.00002354845,5.533726e-7,0.00009775284,0.7013717,0.0002401071,0.08298004,0.00006200455,0.2134724],"study_design_scores_gemma":[0.0004284417,0.0002253922,0.0008428014,0.00002351553,0.00001904559,0.00001469749,0.00008722395,0.9951972,0.0001625619,0.000848104,0.001983804,0.00016722],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00663028,0.00002047094,0.990248,0.000141283,0.0002281023,0.0005401471,7.069852e-8,0.0006070948,0.001584596],"genre_scores_gemma":[0.6213385,0.000003946006,0.3770183,0.00003420488,0.00002467402,0.00007104449,0.00001504772,0.00001082716,0.001483518],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6147082,"threshold_uncertainty_score":0.5607411,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2144283793","doi":"","title":"Bounded Finite State Controllers","year":2003,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":162,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"","keywords":"Bounded function; Mathematical optimization; Observable; Controller (irrigation); State space; Computer science; Mathematics; State (computer science); Vulnerability (computing); Applied mathematics; Control theory (sociology); Algorithm; Control (management); Artificial intelligence; Mathematical analysis","retraction":null,"screen_n_in":null,"score":{"opus":0.01215492872489436,"gpt":0.2210237985235338,"spread":0.2088688697986395,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002465662,0.00008550496,0.00009775798,0.00005526246,0.00008536689,0.0002135307,0.0003859705,0.00002215734,0.0001263219],"category_scores_gemma":[0.0001561756,0.00007308829,0.00003976296,0.0001925857,0.00002628006,0.0002677428,0.00005105325,0.00008489822,0.0004322815],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002610632,"about_ca_system_score_gemma":0.00006187479,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005952946,"about_ca_topic_score_gemma":0.000001587596,"domain_scores_codex":[0.9991403,0.00005839672,0.0001595966,0.0001792616,0.0002076558,0.0002547534],"domain_scores_gemma":[0.9992953,0.0001426761,0.00005279846,0.0003933674,0.00004431416,0.00007157396],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001731864,0.000008968382,0.0002568874,0.000002456966,0.00001862288,0.00001554129,0.0002633493,0.6475833,0.00004052772,0.3494916,0.001401765,0.0009152273],"study_design_scores_gemma":[0.001009674,0.00008927753,0.0001524313,0.000004123676,0.000002681355,0.000008498228,0.00002257489,0.8950768,0.0008418866,0.006477571,0.09610204,0.0002124442],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0004124732,0.000008785731,0.8522407,0.0002042573,0.0002537653,0.0000800361,7.395096e-8,0.0001646523,0.1466352],"genre_scores_gemma":[0.8344108,0.00001094185,0.121856,0.001531228,0.000009143572,0.00000565661,5.881805e-7,0.000008372285,0.04216719],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8339984,"threshold_uncertainty_score":0.5556248,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3081310128","doi":"10.3390/electronics9091363","title":"A Survey of Multi-Task Deep Reinforcement Learning","year":2020,"lang":"en","type":"article","venue":"Electronics","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":160,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Ontario Tech University","funders":"","keywords":"Reinforcement learning; Computer science; Artificial intelligence; Forgetting; Deep learning; Transfer of learning; Task (project management); Active learning (machine learning); Machine learning; Engineering","retraction":null,"screen_n_in":null,"score":{"opus":0.0304471569031292,"gpt":0.2600883677161803,"spread":0.2296412108130511,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000482457,0.0001580715,0.0002293862,0.00005419665,0.00009210123,0.00006429689,0.00090558,0.00006419059,0.00002447252],"category_scores_gemma":[0.0005540651,0.0001629036,0.00006416405,0.0006179765,0.00003133098,0.0002205878,0.0003082385,0.0004390373,0.00008331942],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008095564,"about_ca_system_score_gemma":0.0001952016,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002430933,"about_ca_topic_score_gemma":0.00002018808,"domain_scores_codex":[0.9982764,0.0001551912,0.0003854067,0.0003092022,0.000399631,0.0004741741],"domain_scores_gemma":[0.9989797,0.0001308938,0.000248243,0.0003570538,0.00016011,0.0001239671],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000009056734,0.00001114199,0.002487641,0.00001866621,0.00003323444,0.000002102329,0.0008525922,0.9882504,0.000827658,0.00241575,0.0001024381,0.004989312],"study_design_scores_gemma":[0.0004212394,0.0005545322,0.002223168,0.000006802191,0.000005804539,0.000001268184,0.00001042688,0.9864545,0.001867406,0.000008144958,0.008278321,0.0001683924],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001439241,0.001287803,0.9960057,0.0002646534,0.00007300463,0.0001534681,1.501246e-7,0.0001595842,0.0006163568],"genre_scores_gemma":[0.9895111,0.0002485661,0.009437524,0.000379784,0.00002136213,0.000005022069,0.0000166517,0.00001739952,0.000362598],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9880719,"threshold_uncertainty_score":0.6643016,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W36013382","doi":"10.3390/life12081203","title":"A decision-theoretic approach to task assistance for persons with dementia","year":2005,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":160,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Toronto Rehabilitation Institute; University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Partially observable Markov decision process; Dementia; Task (project management); Personalization; Computer science; Independence (probability theory); Markov decision process; Process (computing); Cognition; Markov process; Artificial intelligence; Human–computer interaction; Cognitive psychology; Risk analysis (engineering); Machine learning; Psychology; Disease; Markov model; Markov chain; Medicine; Engineering; Neuroscience; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.0136587460076659,"gpt":0.2431900778426254,"spread":0.2295313318349596,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002712736,0.0001409064,0.0001300711,0.0000912545,0.0001554059,0.0002212257,0.0008561195,0.00003473452,0.00002589016],"category_scores_gemma":[0.00005751012,0.0001052393,0.00004969545,0.0003292992,0.00003379925,0.0002854029,0.0001384277,0.00007384959,0.00006731624],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005211467,"about_ca_system_score_gemma":0.00004962594,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001349846,"about_ca_topic_score_gemma":0.000004546539,"domain_scores_codex":[0.9986867,0.00001675984,0.0001824683,0.000410979,0.000350201,0.0003528978],"domain_scores_gemma":[0.9989962,0.0001404351,0.00005741918,0.0005855024,0.00008373149,0.0001366863],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002772514,0.0000555437,0.000308744,0.00001356256,0.00004311261,6.841507e-7,0.0006014275,0.7079242,0.0000495343,0.2656686,0.01516386,0.01014302],"study_design_scores_gemma":[0.0005084597,0.0002160959,0.000391867,0.00002179034,0.00001874401,0.000008380868,0.00005065793,0.9281462,0.000162678,0.0003511583,0.06988348,0.0002404545],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0003738822,0.00002255641,0.9611648,0.001060429,0.00006772053,0.0004175068,7.143108e-7,0.0001518791,0.03674053],"genre_scores_gemma":[0.3440697,8.743858e-7,0.6525288,0.0007025991,0.0000321463,0.00005443121,0.000001279916,0.000009688673,0.002600482],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.3436958,"threshold_uncertainty_score":0.4291532,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2136602922","doi":"","title":"Incremental Natural Actor-Critic Algorithms","year":2007,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":157,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Temporal difference learning; Computer science; Convergence (economics); Bellman equation; Mathematical proof; Function approximation; Algorithm; Variance (accounting); Stochastic gradient descent; Function (biology); Rate of convergence; Gradient descent; Artificial intelligence; Mathematical optimization; Mathematics; Artificial neural network","retraction":null,"screen_n_in":null,"score":{"opus":0.01335400574327761,"gpt":0.2696630519567899,"spread":0.2563090462135123,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004145986,0.000112992,0.00008987168,0.00009757355,0.0001014594,0.0001494115,0.0007321856,0.00004052903,0.00008062309],"category_scores_gemma":[0.00004699807,0.00009893506,0.00004740004,0.0002552347,0.00003730446,0.0005194405,0.0002901154,0.0001874801,0.000256017],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008389457,"about_ca_system_score_gemma":0.00002457292,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002857586,"about_ca_topic_score_gemma":0.000004931975,"domain_scores_codex":[0.9987401,0.00001686526,0.0002141265,0.0002287741,0.0004047999,0.000395323],"domain_scores_gemma":[0.9993472,0.0001023362,0.00004122898,0.0003587083,0.00005410389,0.00009643325],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002995878,0.0002255842,0.01542923,0.00007409385,0.0001885078,0.0004439936,0.002590696,0.01973543,0.03406189,0.6049364,0.0138956,0.3083886],"study_design_scores_gemma":[0.0009711609,0.0002446106,0.02680121,0.00002600254,0.000009391599,0.00009641339,0.0002391368,0.9171981,0.03734389,0.0005130997,0.01584345,0.0007135269],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005854929,0.00003209087,0.9669248,0.0002226362,0.0009490399,0.00008605987,9.129278e-8,0.0003007263,0.02562956],"genre_scores_gemma":[0.8444117,0.000001909523,0.1520291,0.0007286769,0.00009362788,9.049734e-7,0.000001791353,0.000006404874,0.002725915],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8974627,"threshold_uncertainty_score":0.4034455,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2785948534","doi":"","title":"NerveNet: Learning Structured Policy with Graph Neural Networks","year":2018,"lang":"en","type":"article","venue":"International Conference on Learning Representations","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":155,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"","keywords":"Reinforcement learning; Concatenation (mathematics); Computer science; Transfer of learning; Artificial intelligence; Graph; Benchmarking; Machine learning; Artificial neural network; Control (management); Theoretical computer science; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.03155445305249521,"gpt":0.3241901345896976,"spread":0.2926356815372024,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000199191,0.0002447878,0.0001743839,0.0004763326,0.000558124,0.000728846,0.001190451,0.00008664132,0.0003598428],"category_scores_gemma":[0.0004420097,0.0002229986,0.00008436919,0.0007366007,0.0002303373,0.0006319091,0.0002624069,0.0007806949,0.00009284297],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009306435,"about_ca_system_score_gemma":0.0001392899,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001766024,"about_ca_topic_score_gemma":0.00003002943,"domain_scores_codex":[0.997707,0.0002255176,0.0003400013,0.0005909106,0.0007317698,0.0004048015],"domain_scores_gemma":[0.998197,0.0002011452,0.0003335754,0.000456176,0.0006725632,0.0001395134],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003055944,0.000014665,0.01517284,0.000001731791,0.00006474299,0.00001168991,0.0009398565,0.8629991,0.0001576712,0.1134301,0.0001022769,0.00707478],"study_design_scores_gemma":[0.0004680232,0.0005168562,0.01805655,0.00004060552,0.000007939104,0.00003467915,0.0002290695,0.9776757,0.0001279989,0.0009978125,0.001591119,0.0002536277],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01786044,0.000006323301,0.9243043,0.003526348,0.0006315684,0.0001796455,0.000001010131,0.0004409599,0.0530494],"genre_scores_gemma":[0.9871295,0.00001581982,0.005333061,0.0003050997,0.0005668472,0.00002630485,0.0000519121,0.00002635578,0.006545114],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.969269,"threshold_uncertainty_score":0.9093618,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2151416233","doi":"","title":"Fitted Q-iteration in continuous action-space MDPs","year":2007,"lang":"en","type":"article","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":149,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Action (physics); Q-learning; Bellman equation; Trajectory; Set (abstract data type); Mathematical optimization; Action selection; Markov decision process; State space; Computer science; State (computer science); Function (biology); Value (mathematics); Space (punctuation); Selection (genetic algorithm); Mathematics; Algorithm; Artificial intelligence; Markov process; Machine learning; Statistics","retraction":null,"screen_n_in":null,"score":{"opus":0.0137757595605324,"gpt":0.2404128600699891,"spread":0.2266371005094567,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005217011,0.0001677522,0.0001749849,0.0002739896,0.0002268912,0.0004198458,0.0009865747,0.0001133806,0.00004201011],"category_scores_gemma":[0.001014889,0.0001859539,0.00006643419,0.0008924998,0.00007896763,0.0005789692,0.0003028178,0.000312473,0.00007901937],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001359155,"about_ca_system_score_gemma":0.00008343826,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003220356,"about_ca_topic_score_gemma":0.0009546096,"domain_scores_codex":[0.9970326,0.001313778,0.0004366607,0.0004358103,0.0003965472,0.000384535],"domain_scores_gemma":[0.9964423,0.001159999,0.0002689856,0.001188625,0.0008198653,0.0001201979],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000260956,0.0005725659,0.03402884,0.0000694524,0.00004901227,0.00004200169,0.02678034,0.01781843,0.02342452,0.7442088,0.001980343,0.1509997],"study_design_scores_gemma":[0.001664039,0.000002274053,0.07118583,0.0006460989,0.00001176881,0.00003463452,0.0003172366,0.7491357,0.1308582,0.001602063,0.04381804,0.0007240969],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05295417,0.00007786212,0.9087112,0.00454998,0.0001914444,0.000206702,6.722555e-7,0.0002433649,0.0330646],"genre_scores_gemma":[0.899999,0.00004884682,0.09045593,0.0001598524,0.00001559217,0.000009687915,0.00003145083,0.00001584571,0.009263756],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8470449,"threshold_uncertainty_score":0.758298,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2955319481","doi":"10.1002/aic.16689","title":"Toward self‐driving processes: A deep reinforcement learning approach to control","year":2019,"lang":"en","type":"article","venue":"AIChE Journal","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":144,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Reinforcement learning; Self driving; Reinforcement; Self-control; Control (management); Computer science; Artificial intelligence; Engineering; Psychology; Social psychology; Transport engineering","retraction":null,"screen_n_in":null,"score":{"opus":0.01407432987377703,"gpt":0.2322518334943767,"spread":0.2181775036205997,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001138623,0.0002533183,0.0003181154,0.0002450039,0.0002879869,0.0007887878,0.001469808,0.00009303768,0.00006086119],"category_scores_gemma":[0.0003866476,0.0002213755,0.0001067042,0.0006346806,0.00001408668,0.0009767221,0.0003393845,0.0008947324,0.000553608],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002161898,"about_ca_system_score_gemma":0.0002938946,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003247775,"about_ca_topic_score_gemma":2.727388e-7,"domain_scores_codex":[0.997352,0.0001332799,0.0005526711,0.0003794606,0.0008958447,0.0006867201],"domain_scores_gemma":[0.9983295,0.000148778,0.0003808956,0.0004385388,0.000360386,0.0003419038],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000008640446,0.00002862933,0.003816576,0.00006669356,0.00007920179,0.000008602614,0.004729819,0.9886008,0.0001800642,0.001373351,0.0002858137,0.0008218081],"study_design_scores_gemma":[0.001118433,0.0004892605,0.0005941306,0.00009198642,0.00002414759,0.0002772099,0.000333345,0.9857123,0.0001446111,0.00005785963,0.01080919,0.0003474725],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004727904,0.00007124669,0.9750943,0.0005210634,0.0004517825,0.0004052233,2.54381e-8,0.0002185559,0.01850989],"genre_scores_gemma":[0.9214317,0.00003249938,0.07533312,0.0009769913,0.000204854,0.00001452373,9.862044e-7,0.00002524547,0.001980126],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9167038,"threshold_uncertainty_score":0.9027432,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2738109916","doi":"10.1109/adconip.2017.7983780","title":"Deep reinforcement learning approaches for process control","year":2017,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":137,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of British Columbia","funders":"","keywords":"Reinforcement learning; Computer science; Process (computing); Controller (irrigation); Artificial neural network; Process control; Control (management); Artificial intelligence; Control system; Control engineering; Function (biology); Deep learning; Control theory (sociology); MIMO; Engineering; Channel (broadcasting)","retraction":null,"screen_n_in":null,"score":{"opus":0.04951332030694924,"gpt":0.2817894084133649,"spread":0.2322760881064156,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004448445,0.0001574064,0.0001824826,0.00005417621,0.0009779142,0.000840867,0.001754096,0.0000629116,0.00002786817],"category_scores_gemma":[0.0004124763,0.0001348024,0.00008312708,0.00004058638,0.0000634754,0.0008750503,0.0002097895,0.0001532778,0.00004735761],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003298042,"about_ca_system_score_gemma":0.0000504732,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007207171,"about_ca_topic_score_gemma":0.000001794894,"domain_scores_codex":[0.9986842,0.00002056788,0.0002518641,0.0003405335,0.0003020117,0.0004008046],"domain_scores_gemma":[0.9985043,0.00009934389,0.000319231,0.0008708144,0.0001148608,0.00009139458],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000008090954,0.000006679224,0.001193602,0.00003128855,0.0000252756,7.909434e-7,0.0002880701,0.925234,0.000006447061,0.06479172,0.00006363242,0.008350421],"study_design_scores_gemma":[0.0009857106,0.0001726643,0.0004783247,0.00001130496,0.00000944029,0.0000021686,0.00006169959,0.9950585,0.0002578341,0.0005769857,0.002192732,0.0001926156],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0001529556,0.0000113598,0.9618305,0.0008147518,0.0002084577,0.0005197524,3.761568e-8,0.0002057317,0.03625648],"genre_scores_gemma":[0.9657788,0.000002606724,0.0272948,0.0001971481,0.00008331164,0.0001135543,0.000002704023,0.00001307282,0.006514054],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9656258,"threshold_uncertainty_score":0.81085,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2128812357","doi":"","title":"Error Propagation for Approximate Policy and Value Iteration","year":2010,"lang":"en","type":"preprint","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":135,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Value (mathematics); Computer science; Mathematical optimization; Algorithm; Mathematics; Machine learning","retraction":null,"screen_n_in":null,"score":{"opus":0.01625520364976859,"gpt":0.2649227238120397,"spread":0.2486675201622711,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001207163,0.000569424,0.0005052505,0.0008113928,0.0003911356,0.001468831,0.001394368,0.0009979499,0.000002966313],"category_scores_gemma":[0.0007681039,0.0005902785,0.0001755852,0.000368411,0.0001062335,0.0006898278,0.001618568,0.001373719,0.00000531463],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004895136,"about_ca_system_score_gemma":0.0008055276,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001838492,"about_ca_topic_score_gemma":0.0001880946,"domain_scores_codex":[0.9968324,0.0001550558,0.0007271074,0.0009951039,0.0005070195,0.0007832728],"domain_scores_gemma":[0.9968113,0.0001952779,0.0007259752,0.001697274,0.0002848264,0.0002853187],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003479986,0.00007894466,0.0006545929,0.0004670693,0.00006154795,0.000007237335,0.000842873,0.4224377,0.006286087,0.5507066,0.0006768217,0.01774578],"study_design_scores_gemma":[0.0003687122,0.0001412042,0.001183825,0.0001136562,0.00003189916,0.00004191695,0.000008990271,0.9719033,0.005304757,0.01793629,0.002394459,0.0005709543],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004994018,0.0001801314,0.9824992,0.007239315,0.0004863556,0.003034835,0.00003622268,0.001170828,0.0003590322],"genre_scores_gemma":[0.2522911,0.0001051237,0.7420113,0.001763101,0.000548629,0.001952829,0.0002192599,0.00009471778,0.001013918],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.5494657,"threshold_uncertainty_score":0.9996549,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2804948070","doi":"","title":"Using Reward Machines for High-Level Task Specification and Decomposition in Reinforcement Learning","year":2018,"lang":"en","type":"article","venue":"International Conference on Machine Learning","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":129,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"","keywords":"Reinforcement learning; Computer science; Task (project management); Decomposition; Artificial intelligence; Human–computer interaction; Engineering","retraction":null,"screen_n_in":null,"score":{"opus":0.09333430035172953,"gpt":0.3548435333796292,"spread":0.2615092330278997,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005974613,0.000226202,0.0002000981,0.0004187428,0.0003053882,0.0003731153,0.0005681012,0.00008287372,0.00008011689],"category_scores_gemma":[0.000330457,0.0002333267,0.0000381215,0.000201728,0.00007441393,0.0005368039,0.0002301561,0.0004673228,0.00003255062],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000188603,"about_ca_system_score_gemma":0.00005438289,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002162561,"about_ca_topic_score_gemma":0.00003057971,"domain_scores_codex":[0.9981607,0.0001355366,0.0004595779,0.0005091034,0.0004370407,0.0002979974],"domain_scores_gemma":[0.9988728,0.0001722133,0.0003412985,0.0002151773,0.0003283149,0.00007024964],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001252948,0.00002815578,0.01054913,0.0000191087,0.00003666681,0.000004320693,0.0008235468,0.6796286,0.01381133,0.2791476,0.00002038995,0.01580594],"study_design_scores_gemma":[0.0008115113,0.0004897515,0.005755191,0.0001301114,0.000006088211,0.00001191976,0.00003789358,0.9876899,0.0006453297,0.003197852,0.0009762649,0.0002482276],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03878467,0.00001254822,0.9538448,0.001216866,0.000495991,0.0002713375,0.000002157976,0.0001187153,0.005252841],"genre_scores_gemma":[0.9404784,0.00003459271,0.05795851,0.0001652802,0.0002059745,0.00002163337,0.00009358234,0.00002031875,0.001021693],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9016938,"threshold_uncertainty_score":0.9514787,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2071814471","doi":"10.1145/1102351.1102472","title":"Bayesian sparse sampling for on-line reward optimization","year":2005,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":122,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Computer science; Machine learning; Bayesian probability; Sampling (signal processing); Action selection; Artificial intelligence; Bayesian optimization; Thompson sampling; Selection (genetic algorithm); Bayesian inference; Mathematical optimization; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.06390024574608347,"gpt":0.3090228942151267,"spread":0.2451226484690433,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002044868,0.00009482022,0.00008949183,0.00007857605,0.0001050655,0.0001308462,0.0004051294,0.00004316707,0.00006110704],"category_scores_gemma":[0.00009804495,0.00008619505,0.00004400756,0.0001499227,0.00001064408,0.0002995117,0.00007143758,0.0000667901,0.00006296292],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004771127,"about_ca_system_score_gemma":0.00002682212,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001802917,"about_ca_topic_score_gemma":0.000001253765,"domain_scores_codex":[0.9991587,0.00001248374,0.0001999891,0.0002349048,0.0001751883,0.0002187774],"domain_scores_gemma":[0.9993139,0.0001159356,0.00006806522,0.0003744966,0.00006703535,0.00006054727],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000003834301,0.00001156692,0.00001825008,0.000003535171,0.000004326756,1.763519e-7,0.00006200455,0.9643478,0.000008761928,0.02522848,0.0006753528,0.009635856],"study_design_scores_gemma":[0.0002683043,0.0001354999,0.000007431322,0.00001067884,0.000002603,0.000001294456,0.000004501995,0.9839454,0.0004338111,0.0001171383,0.01496264,0.0001107329],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00002225471,0.000004492104,0.9897553,0.00252706,0.0001919579,0.0002053299,2.935994e-7,0.0002393114,0.007054018],"genre_scores_gemma":[0.07929439,0.000005886845,0.9159237,0.001270934,0.0001748693,0.0000105669,0.000007168548,0.00001035927,0.00330213],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.07927214,"threshold_uncertainty_score":0.3514932,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2977481643","doi":"10.1609/aaai.v35i12.17276","title":"Improving Sample Efficiency in Model-Free Reinforcement Learning from Images","year":2021,"lang":"en","type":"preprint","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":121,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Robustness (evolution); Computer science; Artificial intelligence; Stability (learning theory); Machine learning; Encoder; Representation (politics); Noise (video); Code (set theory); Image (mathematics)","retraction":null,"screen_n_in":null,"score":{"opus":0.07127289501225326,"gpt":0.2836056152858242,"spread":0.2123327202735709,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","open_science"],"consensus_categories":[],"category_scores_codex":[0.001004913,0.0006026833,0.0007198311,0.0003464766,0.0002590519,0.001234985,0.006937353,0.0003549366,0.00008003823],"category_scores_gemma":[0.003694564,0.0005357641,0.0002971207,0.0007600604,0.0002785183,0.0005825539,0.007959411,0.002202729,0.0000212157],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002768853,"about_ca_system_score_gemma":0.0006024126,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001368372,"about_ca_topic_score_gemma":0.00003726528,"domain_scores_codex":[0.9950778,0.0000568407,0.001467782,0.001345678,0.001311992,0.0007399079],"domain_scores_gemma":[0.9958242,0.0003456781,0.001469649,0.001260296,0.0009707852,0.0001293853],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002481621,0.0000725728,0.0002966763,0.0001345743,0.00002212673,0.000001539708,0.00277178,0.9024622,0.008960551,0.07695018,0.00002363124,0.008279298],"study_design_scores_gemma":[0.00004771079,0.0001026249,0.00005280404,0.0009298519,0.0000202979,6.780675e-7,0.0005437651,0.840656,0.1144794,0.04271866,0.000004452451,0.000443702],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02660413,0.00006107363,0.9663233,0.001035513,0.0007650839,0.000698654,0.000006165523,0.0001626365,0.004343485],"genre_scores_gemma":[0.9547352,0.0001270102,0.04452452,0.0001286741,0.00007668362,0.00006649292,0.0000101102,0.00003510881,0.0002961514],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9281311,"threshold_uncertainty_score":0.9998018,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W359568995","doi":"","title":"A convergent O ( n ) algorithm for off-policy temporal-difference learning with linear function approximation","year":2008,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":120,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Temporal difference learning; Markov decision process; Algorithm; Stochastic approximation; Function (biology); Reinforcement learning; Stochastic gradient descent; Markov process; Mathematics; Quadratic equation; Approximation algorithm; Norm (philosophy); Computer science; Function approximation; Mathematical optimization; Applied mathematics; Artificial intelligence; Statistics; Artificial neural network","retraction":null,"screen_n_in":null,"score":{"opus":0.02640782854945645,"gpt":0.2470170740798348,"spread":0.2206092455303784,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001814665,0.000168045,0.0001626731,0.0001541893,0.0003228672,0.00006627512,0.0003262472,0.00006531212,0.00001912699],"category_scores_gemma":[0.00006959923,0.0001335624,0.00005151051,0.0004188903,0.00005308939,0.0003923725,0.00008964076,0.0001664494,0.00005295117],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007055081,"about_ca_system_score_gemma":0.0001779047,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004798926,"about_ca_topic_score_gemma":0.000001435478,"domain_scores_codex":[0.9986966,0.00004415103,0.0002529185,0.0003455084,0.0003602682,0.000300558],"domain_scores_gemma":[0.9991305,0.00007789039,0.0001778691,0.0003132667,0.0002092476,0.00009122711],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004814753,0.00009628459,0.006016062,0.00006157948,0.00008575082,0.000008028179,0.001479441,0.6766451,0.0003322593,0.01790076,0.0007215899,0.296605],"study_design_scores_gemma":[0.0006507232,0.0006401619,0.001450273,0.00001361199,0.000006302272,0.00002383586,0.00003284213,0.988685,0.0005460291,0.00008043269,0.007679178,0.0001915766],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001831732,0.0000159449,0.995727,0.0002553173,0.0001451855,0.0004144938,3.886671e-7,0.0003927519,0.001217196],"genre_scores_gemma":[0.1858481,0.00004416417,0.7929378,0.0003307465,0.0002120807,0.00008570671,0.00003974153,0.00002366077,0.02047802],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.31204,"threshold_uncertainty_score":0.5446516,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2144794447","doi":"","title":"Bayes-Adaptive POMDPs","year":2007,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":112,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université Laval; McGill University","funders":"","keywords":"Partially observable Markov decision process; Reinforcement learning; Computer science; Markov decision process; Bellman equation; Artificial intelligence; Bayesian probability; Machine learning; Bayes' theorem; Markov process; Markov chain; Mathematical optimization; Markov model; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.01648638832331079,"gpt":0.2496534226924639,"spread":0.2331670343691531,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004021169,0.00007206541,0.0000627179,0.00007013811,0.00006101012,0.00006755251,0.0005549037,0.00003370327,0.00005390037],"category_scores_gemma":[0.00002369234,0.00006167231,0.00003183405,0.0002348636,0.00001492526,0.0002768964,0.0001247725,0.00009398271,0.0004751258],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003904704,"about_ca_system_score_gemma":0.00002412362,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001061543,"about_ca_topic_score_gemma":0.000002373072,"domain_scores_codex":[0.9991756,0.00001187446,0.0001405978,0.0001769055,0.000234447,0.0002605447],"domain_scores_gemma":[0.9993888,0.00009925981,0.00004017972,0.0003514557,0.00004622371,0.0000740461],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000004483453,0.0000272662,0.001306717,0.000003115499,0.00002408974,0.00003069538,0.0007057164,0.02383599,0.0006030916,0.9552886,0.002481435,0.01568879],"study_design_scores_gemma":[0.0007123906,0.0005751841,0.01946405,0.00001760776,0.000008943068,0.00004644213,0.0001721828,0.8578063,0.02553957,0.00369228,0.09126528,0.0006997209],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0001938093,0.000006522531,0.7602589,0.0001380507,0.0001999228,0.00004703738,1.699853e-8,0.0003163817,0.2388394],"genre_scores_gemma":[0.525574,0.000001710581,0.4690603,0.0005419122,0.00004357845,6.09394e-7,2.079243e-7,0.000004321103,0.004773354],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9515963,"threshold_uncertainty_score":0.6106939,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2027648864","doi":"10.2991/agi.2010.22","title":"GQ( ): A general gradient algorithm for temporal-difference prediction learning with eligibility traces","year":2010,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":109,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Algorithm; Temporal difference learning; Artificial intelligence; Reinforcement learning","retraction":null,"screen_n_in":null,"score":{"opus":0.01308984256616065,"gpt":0.2579115051959552,"spread":0.2448216626297946,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004505109,0.0001826239,0.00017067,0.00007936605,0.0002634327,0.000246102,0.0005579583,0.000086987,0.00001960147],"category_scores_gemma":[0.00007276851,0.0001352367,0.00006044418,0.0002168156,0.00007928703,0.0003920875,0.0001034573,0.0003940005,0.00001030021],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003209908,"about_ca_system_score_gemma":0.00007800527,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007327184,"about_ca_topic_score_gemma":0.00003708509,"domain_scores_codex":[0.9984556,0.00004132954,0.0002747715,0.0005122874,0.0003545904,0.0003614297],"domain_scores_gemma":[0.9989601,0.0001071669,0.0001454339,0.0004938797,0.0001677224,0.0001257483],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005295107,0.0003826831,0.1630436,0.00009862524,0.0001323855,0.00001023708,0.002571864,0.4418991,0.008004176,0.03400784,0.001271263,0.3485253],"study_design_scores_gemma":[0.0004507315,0.0006919646,0.01398781,0.000008676298,0.000008208644,0.00001448373,0.0000243212,0.9795952,0.001496789,0.0002739739,0.003266527,0.0001813111],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1315181,0.000004918113,0.8664453,0.0001775486,0.0004156128,0.0003806311,0.000001546308,0.0003561436,0.0007002243],"genre_scores_gemma":[0.4479979,0.000003115518,0.5469892,0.00005634071,0.0001122827,0.00005511564,0.00001292325,0.00001063015,0.004762543],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.5376961,"threshold_uncertainty_score":0.551479,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2158738729","doi":"","title":"Regularized Policy Iteration","year":2008,"lang":"en","type":"article","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":108,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University; University of Alberta","funders":"","keywords":"Reinforcement learning; Mathematical optimization; Regularization (linguistics); Reproducing kernel Hilbert space; Temporal difference learning; Computer science; Bellman equation; Markov decision process; Kernel (algebra); Residual; Hilbert space; Parametric statistics; Convergence (economics); Function approximation; Rate of convergence; Algorithm; Mathematics; Artificial intelligence; Markov process; Artificial neural network","retraction":null,"screen_n_in":null,"score":{"opus":0.01361379034717296,"gpt":0.2341841148069613,"spread":0.2205703244597883,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004707746,0.0003094737,0.0002998483,0.0006595723,0.0005060619,0.0003137203,0.001314647,0.0002315038,0.00001861109],"category_scores_gemma":[0.0003584282,0.0003197571,0.000161207,0.001150244,0.0001027783,0.001056909,0.0004395104,0.000400826,0.00009018645],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000498026,"about_ca_system_score_gemma":0.0005043736,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001971768,"about_ca_topic_score_gemma":0.00006714044,"domain_scores_codex":[0.9974926,0.0001566324,0.0005062401,0.0005044159,0.0005875275,0.0007525987],"domain_scores_gemma":[0.9977376,0.00009501584,0.0002518614,0.001471461,0.0001563304,0.0002877525],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004563012,0.0001717061,0.01674935,0.00003131398,0.00007035593,0.0002627503,0.001766233,0.3087007,0.02135608,0.6291784,0.008296905,0.01337061],"study_design_scores_gemma":[0.0005910139,0.0001541286,0.01993431,0.00002518357,0.000007102347,0.0004419751,0.000009519779,0.9604462,0.008672397,0.002380203,0.006897449,0.0004405379],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01190158,0.0001304425,0.9768935,0.006540798,0.0001533279,0.0004497047,0.000002049966,0.001467012,0.002461625],"genre_scores_gemma":[0.5925515,0.000128722,0.3984527,0.003250908,0.0002097861,0.0001312214,0.00001464619,0.00003765041,0.005222841],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6517456,"threshold_uncertainty_score":0.9999254,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1758031947","doi":"10.48550/arxiv.1206.3285","title":"Dyna-Style Planning with Linear Function Approximation and Prioritized Sweeping","year":2012,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":107,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Computer science; Bellman equation; Function (biology); Limit (mathematics); Mathematical optimization; Function approximation; State (computer science); Linear approximation; Linear programming; Artificial intelligence; Algorithm; Mathematics; Nonlinear system; Artificial neural network","retraction":null,"screen_n_in":null,"score":{"opus":0.0529436338934253,"gpt":0.1830294782193934,"spread":0.1300858443259681,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002025758,0.0001068703,0.0001004913,0.0001114037,0.0001835884,0.00005720959,0.0001989646,0.00005277196,0.000005460985],"category_scores_gemma":[0.00002028863,0.0001096309,0.00002114767,0.0003752896,0.00004268524,0.001274338,0.0001301308,0.0001328786,0.00002558579],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005572888,"about_ca_system_score_gemma":0.00001931882,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008326833,"about_ca_topic_score_gemma":5.329262e-7,"domain_scores_codex":[0.9992925,0.00004635317,0.00008598977,0.0002470389,0.00007362192,0.0002544982],"domain_scores_gemma":[0.9994273,0.00005992224,0.0001020354,0.0002580224,0.00004992411,0.0001027593],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004319753,0.00001975267,0.04204223,0.00002822346,0.00003088632,0.00001235672,0.0007536652,0.8852861,0.0001239684,0.07096841,0.00001561814,0.0006755872],"study_design_scores_gemma":[0.0005733097,0.00008765526,0.01010059,0.00003392547,0.00002437418,0.000007861775,0.0001628294,0.9881592,0.00006875463,0.0001210743,0.0004977189,0.0001627511],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1747361,0.00002108137,0.8230492,0.0000172856,0.00009858234,0.00009571839,1.182121e-7,0.0001461194,0.001835802],"genre_scores_gemma":[0.9878357,0.000007089324,0.01149435,0.00005276211,0.00005849269,2.996432e-7,0.000003155426,0.000007728426,0.0005404237],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8130996,"threshold_uncertainty_score":0.4470619,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2593237273","doi":"10.1609/aaai.v32i1.11631","title":"Multi-Step Reinforcement Learning: A Unifying Algorithm","year":2018,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":107,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; Alberta Innovates; DeepMind","keywords":"Reinforcement learning; Computer science; Backup; Algorithm; TRACE (psycholinguistics); Sampling (signal processing); Focus (optics); Monte Carlo method; Importance sampling; Artificial intelligence; Machine learning; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.09247718432083181,"gpt":0.3136361361909427,"spread":0.2211589518701109,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007418666,0.000309026,0.0002912106,0.0001948037,0.000495519,0.0004152124,0.002859985,0.0001209567,0.0001257517],"category_scores_gemma":[0.000633862,0.000242286,0.000141166,0.0008994911,0.0005071139,0.0005502939,0.0009153195,0.0005507913,0.0004037424],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009122324,"about_ca_system_score_gemma":0.0001222704,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005174801,"about_ca_topic_score_gemma":0.000003825049,"domain_scores_codex":[0.9972612,0.00002945142,0.0007430391,0.0005864651,0.0008140761,0.0005658161],"domain_scores_gemma":[0.9975198,0.00008706097,0.0006487655,0.0004544693,0.001158748,0.0001311774],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005289292,0.0001479687,0.0003982802,0.0000670603,0.00006339402,0.000001475817,0.004897086,0.02834892,0.02942686,0.7260391,0.0002988776,0.2102581],"study_design_scores_gemma":[0.00004546647,0.0004822668,0.00007046048,0.0001787651,0.00001013259,0.000004298136,0.0004791254,0.8288441,0.1661968,0.002635838,0.0008069917,0.0002457839],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002891157,0.000007234223,0.9794903,0.001036294,0.0006558921,0.0004158699,4.018624e-7,0.0002000973,0.01530271],"genre_scores_gemma":[0.9133315,0.00002413619,0.08332956,0.0002385959,0.0001348117,0.00001988864,5.478779e-7,0.00001963346,0.002901307],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9104404,"threshold_uncertainty_score":0.9880135,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null}]}