{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00179616,0.001154558,0.001409557,0.002675164,0.0004074779,0.002165541,0.001558592,0.001961206,0.00389698],"category_scores_gemma":[0.003938644,0.0006143135,0.0007157455,0.00442322,0.00113277,0.003142055,0.001051206,0.002958983,0.001991165],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001330001,"about_ca_system_score_gemma":0.002184543,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002105322,"about_ca_topic_score_gemma":0.001865717,"domain_scores_codex":[0.9994215,0.0001382155,0.00008101267,0.00014846,0.0001772007,0.00003373698],"domain_scores_gemma":[0.996796,0.002390386,0.000168001,0.00007519202,0.000472509,0.00009806215],"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":[0.00007687198,0.0001449571,0.0003916822,0.02293243,0.0001241256,0.0001457375,0.0002302625,0.003641831,0.0005828555,0.0431855,0.01576897,0.9127749],"study_design_scores_gemma":[0.00003041949,0.0002574795,0.001106333,0.01509455,0.0001905615,0.0009705718,0.0002936817,0.002559174,0.0006364376,0.03512603,0.9436439,0.0000908437],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.0001791507,0.9949846,0.002204542,0.0006094696,0.0001689209,0.00000709577,0.00001005789,0.00001718403,0.001819003],"genre_scores_gemma":[0.002637474,0.99426,0.001881602,0.00036713,0.000347138,0.00001608421,0.00002534957,0.000008224983,0.0004569902],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.00389698,"threshold_uncertainty_score":0.01303673,"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."}}