{"id":"W4388483498","doi":"10.1109/ase56229.2023.00121","title":"An Intentional Forgetting-Driven Self-Healing Method for Deep Reinforcement Learning Systems","year":2023,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Forgetting; Reinforcement learning; Convergence (economics); Computer science; Adaptation (eye); Reinforcement; Artificial intelligence; Cognitive psychology; Engineering; 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":[],"consensus_categories":[],"category_scores_codex":[0.001388903,0.000211104,0.0002457647,0.0003069564,0.0004861941,0.000475961,0.000960285,0.00009992009,0.00001411177],"category_scores_gemma":[0.0001911212,0.000201514,0.0001341706,0.0006075993,0.00001467774,0.0007108888,0.0003114577,0.0002375827,0.000131771],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001321691,"about_ca_system_score_gemma":0.00007502726,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000342209,"about_ca_topic_score_gemma":0.000001682172,"domain_scores_codex":[0.997539,0.0001578504,0.0005693902,0.000522044,0.0005652551,0.0006464734],"domain_scores_gemma":[0.9983258,0.0004526713,0.0002546803,0.0005145029,0.0002835607,0.0001687685],"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":[0.000003016686,0.000008771099,0.0003467745,0.00007579326,0.00004797367,0.000002297498,0.0007673734,0.9009835,0.0003643484,0.09592783,0.0004754422,0.0009969312],"study_design_scores_gemma":[0.0003989926,0.0003170043,0.00008124307,0.00003705038,0.00001464227,0.000008061917,0.0005903228,0.9904234,0.0002251849,0.0001872855,0.007460513,0.0002563418],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0001491431,0.00001597494,0.9940841,0.0002720508,0.0008620894,0.0006139766,2.370518e-7,0.001977145,0.002025256],"genre_scores_gemma":[0.2708358,0.00001252804,0.7227693,0.0002084439,0.0002290777,0.0001739652,0.00008834493,0.00004037251,0.005642137],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2713148,"threshold_uncertainty_score":0.8217501,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02687934710323557,"score_gpt":0.3167402699521178,"score_spread":0.2898609228488823,"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."}}