{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009223534,0.0002192465,0.0003263693,0.0003257226,0.0001613815,0.000188485,0.002136129,0.00006912412,0.0000448867],"category_scores_gemma":[0.001138179,0.0001826376,0.0001312023,0.002114766,0.000197107,0.0003357576,0.0008671129,0.0003560454,0.0001689732],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005699352,"about_ca_system_score_gemma":0.0001352345,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004507347,"about_ca_topic_score_gemma":0.00000180544,"domain_scores_codex":[0.9975744,0.00002869447,0.0006726205,0.0004666425,0.0007855231,0.0004721013],"domain_scores_gemma":[0.998212,0.0001349083,0.0005060469,0.0003294535,0.0006997765,0.0001178358],"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":[0.00003846505,0.00003471047,0.001184089,0.00006685626,0.00002360191,9.798515e-7,0.005964089,0.4709082,0.1114451,0.3838189,0.00008442326,0.02643051],"study_design_scores_gemma":[0.00002388029,0.0002471151,0.0001479463,0.0002702616,0.000009345761,0.000001979135,0.0006981958,0.7999372,0.1888696,0.009533662,0.00008797129,0.0001727742],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8082442,0.000004762935,0.1815467,0.001837472,0.0005623152,0.0005591087,0.000001360284,0.0003052412,0.006938857],"genre_scores_gemma":[0.9928456,0.00001345343,0.00643884,0.0001106918,0.00004707847,0.00001158155,3.35897e-7,0.0000177757,0.0005146569],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3742853,"threshold_uncertainty_score":0.7447743,"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."}}