{"id":"W3170371439","doi":"10.1609/aaai.v36i7.20758","title":"Control-Oriented Model-Based Reinforcement Learning with Implicit Differentiation","year":2022,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Vector Institute; University of Toronto; Université de Montréal","funders":"Compute Canada","keywords":"Reinforcement learning; Bellman equation; Computer science; Context (archaeology); Function (biology); Task (project management); Class (philosophy); Value (mathematics); Likelihood function; Maximum likelihood; Reinforcement; Control (management); Artificial intelligence; Mathematical optimization; Machine learning; Estimation theory; Mathematics; Statistics; Economics; Algorithm","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005529324,0.0002654507,0.0002775447,0.0001947642,0.0007261955,0.0002100048,0.001823563,0.00004680949,0.0001007017],"category_scores_gemma":[0.0001738368,0.0002042991,0.0001130726,0.0007932207,0.0001455981,0.0003318681,0.000497819,0.0006325038,0.00001584567],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001679205,"about_ca_system_score_gemma":0.0001805098,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002266539,"about_ca_topic_score_gemma":0.00000123493,"domain_scores_codex":[0.9972706,0.00004160403,0.0005912123,0.0005050915,0.00115396,0.0004375215],"domain_scores_gemma":[0.998147,0.00009074395,0.0007406122,0.0003677878,0.0005671874,0.0000866734],"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":[0.0001049973,0.00004678458,0.0003617932,0.00001401428,0.00001492253,2.19264e-7,0.0003722357,0.6246036,0.007002605,0.3658809,0.00001988982,0.001577991],"study_design_scores_gemma":[0.0001246889,0.0007809984,0.00008664717,0.00005797237,0.00002119988,0.000002063063,0.0002109977,0.9391222,0.05488533,0.004401106,0.00007649296,0.0002302966],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03704292,0.000002981341,0.9571276,0.001463551,0.0001852493,0.0005469458,0.000001304405,0.0001493088,0.003480103],"genre_scores_gemma":[0.9956216,0.000002337852,0.003377889,0.0003658642,0.00001925195,0.0001174169,0.000002808745,0.00001919109,0.0004736881],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9585786,"threshold_uncertainty_score":0.8331077,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03202434377017425,"score_gpt":0.2550913103624891,"score_spread":0.2230669665923148,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}