{"id":"W3042725540","doi":"10.48550/arxiv.2007.07011","title":"Lifelong Policy Gradient Learning of Factored Policies for Faster Training Without Forgetting","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"","keywords":"Forgetting; Lifelong learning; Computer science; Train; Task (project management); Reuse; Process (computing); Function (biology); Variety (cybernetics); Training (meteorology); Control (management); Artificial intelligence; Machine learning; Cognitive psychology; Economics; Political science; Engineering; Psychology; Management","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003397954,0.0003549647,0.000550307,0.0004940496,0.0002743111,0.0001602289,0.00124777,0.0002088801,0.000005924201],"category_scores_gemma":[0.0004092674,0.0004157557,0.0003595055,0.0006595263,0.0001152535,0.00035982,0.001102904,0.0006638023,0.0000087142],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001603575,"about_ca_system_score_gemma":0.0003986042,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001663777,"about_ca_topic_score_gemma":0.00001392652,"domain_scores_codex":[0.9978579,0.000174931,0.0003770278,0.0009167654,0.0001467797,0.0005266113],"domain_scores_gemma":[0.998118,0.0002589865,0.0006691678,0.0005038338,0.0001987865,0.0002512507],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00009120504,0.00005891102,0.01316942,0.0003957588,0.0002867749,0.00003968427,0.03918459,0.4671743,0.0009026711,0.4681363,0.00007866378,0.01048171],"study_design_scores_gemma":[0.0009706067,0.0001626968,0.00319258,0.0002077882,0.00005842038,0.000003742064,0.003526547,0.9744799,0.0002894664,0.01335176,0.003164247,0.0005922893],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1684622,0.00002232528,0.8281529,0.0005549111,0.0002544165,0.0003682036,0.00001015444,0.0002714765,0.001903466],"genre_scores_gemma":[0.9866654,0.00002315091,0.01178477,0.0002585287,0.0001989797,0.000002567682,0.00002207817,0.00003694166,0.001007629],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8182032,"threshold_uncertainty_score":0.9998294,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1563627802819452,"score_gpt":0.2347216323033455,"score_spread":0.0783588520214003,"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."}}