{"id":"W3089723243","doi":"","title":"Learning Intrinsic Rewards as a Bi-Level Optimization Problem","year":2020,"lang":"en","type":"article","venue":"Uncertainty in Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Intrinsic motivation; Mathematical optimization; Mathematics; Psychology; Social psychology","routes":{"ca_aff":true,"ca_fund":false,"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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005935945,0.0002674374,0.0002884361,0.0002252524,0.000192281,0.0003696927,0.001226044,0.0001369031,0.0001636265],"category_scores_gemma":[0.001416976,0.0002791107,0.00008541073,0.001751169,0.0001121979,0.0005606853,0.0004332732,0.0006762156,0.0006226582],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001740083,"about_ca_system_score_gemma":0.0002269295,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002613764,"about_ca_topic_score_gemma":0.00003017375,"domain_scores_codex":[0.997175,0.0002055981,0.0007975627,0.0006971014,0.0005722038,0.0005524899],"domain_scores_gemma":[0.9986991,0.0002288305,0.0002481251,0.0003980274,0.0002187252,0.0002071441],"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":[0.00002979709,0.00002649894,0.0001391639,0.0000195082,0.000008003498,0.00002526094,0.004339095,0.8736681,0.0001548751,0.05331301,0.00003526275,0.06824138],"study_design_scores_gemma":[0.00005195508,0.0003956192,0.0000174925,0.00006140299,0.00000448466,0.000005598848,0.000560008,0.9907634,0.002118227,0.004667889,0.001034352,0.0003196142],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001744108,0.00002618891,0.9894383,0.004360224,0.000252179,0.0004057195,7.745625e-7,0.0003566405,0.003415862],"genre_scores_gemma":[0.8751268,0.00006660161,0.1231025,0.00124528,0.0001542735,0.00003237151,0.00001319792,0.00002597584,0.000232998],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8733827,"threshold_uncertainty_score":0.9999661,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06927971827994504,"score_gpt":0.2923596390382754,"score_spread":0.2230799207583304,"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."}}