{"id":"W4400373827","doi":"10.48550/arxiv.2407.01704","title":"Weight Clipping for Deep Continual and Reinforcement Learning","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Muscle activation and electromyography studies","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alliance de recherche numérique du Canada; Alberta Innovates; Alberta Machine Intelligence Institute; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Clipping (morphology); Reinforcement; Computer science; Artificial intelligence; Psychology; Social psychology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002642419,0.0009636804,0.0006429621,0.0004291647,0.0005347871,0.0007856666,0.001475518,0.001122698,0.002452246],"category_scores_gemma":[0.01422755,0.0004588438,0.0003832406,0.000409571,0.001598861,0.001828303,0.00155244,0.002857941,0.0004012602],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009085464,"about_ca_system_score_gemma":0.0008600746,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002512495,"about_ca_topic_score_gemma":0.002183916,"domain_scores_codex":[0.9992846,0.0002257956,0.00004714823,0.000173553,0.0001889874,0.00008002626],"domain_scores_gemma":[0.996455,0.001873852,0.0003876456,0.0006733133,0.0003913571,0.0002188709],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0003873605,0.0003397174,0.002450315,0.0002051309,0.0001194,0.0002030463,0.0002538932,0.7285638,0.01275805,0.02977859,0.003768748,0.2211719],"study_design_scores_gemma":[0.00002218639,0.0001007114,0.0002501557,0.00001383581,0.000008346042,0.00002914744,0.00001125973,0.9812004,0.002608513,0.01504544,0.0006990511,0.00001095637],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05724517,0.0005004274,0.9382918,0.0004184602,0.0001033456,0.00006707779,0.00005416874,0.001193683,0.002125873],"genre_scores_gemma":[0.8744901,0.0002272121,0.1223963,0.0002745522,0.00007684518,0.0001608274,0.000107387,0.0001651467,0.002101736],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002642419,"threshold_uncertainty_score":0.01397461,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02980102731404241,"score_gpt":0.1724299642763628,"score_spread":0.1426289369623204,"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."}}