{"id":"W4417239585","doi":"10.1038/s41467-025-66009-y","title":"Discovery of the reward function for embodied reinforcement learning agents","year":2025,"lang":"en","type":"article","venue":"Nature Communications","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Victoria","funders":"National Key Research and Development Program of China; State Key Laboratory of Industrial Control Technology; State Key Laboratory of Mechanical Transmissions; Huazhong University of Science and Technology; National Natural Science Foundation of China","keywords":"Embodied cognition; Reinforcement learning; Regret; Maximization; Adaptability; Cognition; Function (biology); Cognitive robotics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001030475,0.0005498978,0.0007545646,0.000375141,0.000394155,0.0009040797,0.0007552023,0.001161319,0.001350577],"category_scores_gemma":[0.004398306,0.0004415218,0.0004769498,0.0002037405,0.001252528,0.000984089,0.001427353,0.001326783,0.0003099798],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008194904,"about_ca_system_score_gemma":0.00106557,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001105163,"about_ca_topic_score_gemma":0.0009719761,"domain_scores_codex":[0.9996804,0.00009826229,0.00001780739,0.00006064761,0.00009064173,0.00005220851],"domain_scores_gemma":[0.9990869,0.0005141536,0.0001325352,0.00006069998,0.0001224683,0.00008333506],"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":[0.00004805197,0.00003994994,0.0006860819,0.00007324508,0.00002415083,0.0001092478,0.0001000044,0.85861,0.005563372,0.1142013,0.0006415738,0.01990304],"study_design_scores_gemma":[0.000007022646,0.00001540901,0.00003937505,0.000006811101,0.000003127063,0.00001239647,0.000005534973,0.9823569,0.0006024273,0.01651941,0.0004270434,0.000004582876],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02670989,0.0001467571,0.9690275,0.0003296825,0.00002457859,0.00002919657,0.00001529605,0.0001287469,0.003588458],"genre_scores_gemma":[0.7922278,0.0002733859,0.2036972,0.0001331924,0.00002382258,0.0001333251,0.00004129242,0.00008521487,0.003384716],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.001350577,"threshold_uncertainty_score":0.005945861,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02517842891644697,"score_gpt":0.3064949102456547,"score_spread":0.2813164813292078,"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."}}