{"id":"W3117215073","doi":"10.1613/jair.1.13673","title":"Towards Continual Reinforcement Learning: A Review and Perspectives","year":2022,"lang":"en","type":"review","venue":"Journal of Artificial Intelligence Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":180,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; McGill University","funders":"Canada Excellence Research Chairs, Government of Canada; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Computer science; Scope (computer science); Artificial intelligence; Bridging (networking); Taxonomy (biology); Function (biology); Management science; 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":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.01167099,0.0004019499,0.001920793,0.001212323,0.0005446997,0.0005758179,0.002878048,0.000173706,0.0008247282],"category_scores_gemma":[0.004848099,0.0003204127,0.000599543,0.00201187,0.0003637585,0.0005506512,0.001777283,0.004585868,0.0000898108],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000670457,"about_ca_system_score_gemma":0.001834283,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001694632,"about_ca_topic_score_gemma":7.451476e-7,"domain_scores_codex":[0.991866,0.002030564,0.002180656,0.0005497446,0.002603675,0.0007693485],"domain_scores_gemma":[0.994913,0.001465323,0.001499777,0.0006523895,0.001139064,0.0003304786],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000006593099,0.00004404091,1.560825e-7,0.006680058,0.0001705447,0.0001121582,0.001133247,0.002921728,2.790719e-7,0.02010213,0.000697781,0.9681313],"study_design_scores_gemma":[0.00001913925,0.001293918,1.013555e-7,0.008857085,0.0002132788,0.0003924475,0.00108381,0.003642684,0.000004018696,0.0005308807,0.983661,0.0003016187],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[1.006542e-7,0.7782069,0.21892,0.0005247617,0.0002594229,0.0006263192,5.310488e-7,0.00002235593,0.00143958],"genre_scores_gemma":[0.00002268204,0.9961339,0.00233357,0.00003572108,0.0002699145,0.00003875874,0.000002744795,0.00003634514,0.001126314],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9829633,"threshold_uncertainty_score":0.9999248,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3537488428002021,"score_gpt":0.4758490705427069,"score_spread":0.1221002277425048,"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."}}