{"id":"W4415196822","doi":"10.1016/j.neunet.2025.108202","title":"Chaos-based reinforcement learning with TD3","year":2025,"lang":"en","type":"article","venue":"Neural Networks","topic":"Neural Networks and Reservoir Computing","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Institute of Aging; International Research Center for Neurointelligence, University of Tokyo; New Energy and Industrial Technology Development Organization; Japan Science and Technology Agency; Japan Society for the Promotion of Science; University of Tokyo","keywords":"Reinforcement learning; Chaotic; Range (aeronautics); Q-learning; Reinforcement; Learning classifier system; Stability (learning theory); Action (physics)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001912708,0.000230101,0.0002219687,0.000107637,0.0004134125,0.0002708911,0.0008438439,0.00008444116,0.00001045028],"category_scores_gemma":[0.00001162345,0.0001691475,0.00008757551,0.000921386,0.00005134263,0.0002204646,0.0003577117,0.0005872883,0.000004778655],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003680562,"about_ca_system_score_gemma":0.0000451925,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001735786,"about_ca_topic_score_gemma":0.000009224961,"domain_scores_codex":[0.998301,0.00009689149,0.0002664815,0.0004815352,0.0002660585,0.0005880562],"domain_scores_gemma":[0.9990281,0.0001837416,0.0001139924,0.0004893578,0.0000783283,0.0001064498],"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":[0.00002358542,0.00001325049,0.001577963,0.00001129003,0.0000138691,0.00003658331,0.00001319129,0.9538904,0.00001167375,0.003100732,0.001098419,0.04020907],"study_design_scores_gemma":[0.0005109364,0.0001930175,0.0006610693,0.0001062689,0.000007148894,0.000006507477,0.000003964484,0.9926965,0.00008902522,0.00007182666,0.005455384,0.000198348],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01323052,0.0003736897,0.9782979,0.002640492,0.0007949386,0.0002429264,3.889616e-8,0.0004435012,0.003976015],"genre_scores_gemma":[0.9928453,0.00001503136,0.003001838,0.002605663,0.0002208894,0.00001595163,0.000003859028,0.00001258734,0.001278912],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9796147,"threshold_uncertainty_score":0.6897634,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007393415344419196,"score_gpt":0.2231342666939981,"score_spread":0.2157408513495788,"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."}}