{"id":"W4382202736","doi":"10.1609/aaai.v37i10.26371","title":"Learning to Shape Rewards Using a Game of Two Partners","year":2023,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs; University of Alberta; UK Research and Innovation; Alberta Machine Intelligence Institute; Compute Canada; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Computer science; Task (project management); Function (biology); Convergence (economics); Construct (python library); Markov decision process; Artificial intelligence; State (computer science); Domain (mathematical analysis); Machine learning; Markov process; Algorithm; Mathematics; Engineering","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001782895,0.001248649,0.0008365549,0.0003421887,0.0007252915,0.001009763,0.001601634,0.001730065,0.004885106],"category_scores_gemma":[0.006974917,0.0005172927,0.0006844492,0.0002411776,0.001949827,0.001677501,0.002176572,0.001761664,0.0005543504],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00103519,"about_ca_system_score_gemma":0.001570906,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002941157,"about_ca_topic_score_gemma":0.002987669,"domain_scores_codex":[0.9989791,0.0004808117,0.00004521236,0.0001863958,0.0001765999,0.0001319105],"domain_scores_gemma":[0.9972474,0.001755245,0.000274692,0.0002273232,0.0001649928,0.0003302796],"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":[0.0002713693,0.0002206107,0.001456254,0.00009497629,0.00007238526,0.0003447041,0.0003812924,0.883523,0.004889905,0.0694508,0.001248434,0.03804613],"study_design_scores_gemma":[0.00004794684,0.00008520624,0.0001044282,0.00001021002,0.000009220837,0.00003715593,0.00002786021,0.9792367,0.0007736336,0.01864452,0.001011521,0.0000116119],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09286436,0.0001195796,0.8928725,0.0007025832,0.00004799591,0.0003020561,0.0000574662,0.0005536533,0.01247975],"genre_scores_gemma":[0.8495136,0.00008640532,0.1435293,0.0002110336,0.00001924049,0.0003806774,0.00005069852,0.00006677541,0.00614219],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004885106,"threshold_uncertainty_score":0.01634228,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1508422749642497,"score_gpt":0.372797897068085,"score_spread":0.2219556221038352,"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."}}