{"id":"W4320855027","doi":"10.48550/arxiv.2302.06548","title":"Automatic Noise Filtering with Dynamic Sparse Training in Deep Reinforcement Learning","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Evolutionary Algorithms and Applications","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs; University of Alberta; Alberta Machine Intelligence Institute; Compute Canada; Nederlandse Organisatie voor Wetenschappelijk Onderzoek; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Computer science; Noise (video); Focus (optics); Task (project management); Artificial intelligence; Margin (machine learning); Robot; Code (set theory); Machine learning; Deep learning; Pattern recognition (psychology); Set (abstract data type)","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":[0.001347906,0.001000433,0.0009209219,0.0002802297,0.0003571722,0.0006544346,0.001627526,0.001046022,0.001962947],"category_scores_gemma":[0.005555057,0.0004835323,0.0004346202,0.0003175297,0.001232715,0.001151113,0.001251848,0.002436778,0.0004284716],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001052798,"about_ca_system_score_gemma":0.001566639,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007711299,"about_ca_topic_score_gemma":0.009594936,"domain_scores_codex":[0.9995073,0.0001486741,0.00002303066,0.0001209039,0.0001201117,0.00008002305],"domain_scores_gemma":[0.9984344,0.0009585708,0.0001541961,0.0001664099,0.0001905268,0.00009581175],"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":[0.0001086166,0.00009264244,0.0008877908,0.00006240943,0.0000323392,0.00005199452,0.00005477076,0.9295664,0.002273061,0.009036077,0.001660532,0.05617335],"study_design_scores_gemma":[0.00001010501,0.00001593515,0.00003259645,0.000003718025,0.00000249246,0.000004497288,0.000002296806,0.9961447,0.0004305741,0.003155199,0.0001956507,0.000002383047],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02566059,0.000287473,0.9697068,0.0003672935,0.00006034399,0.00005239563,0.00008205343,0.001791867,0.001991153],"genre_scores_gemma":[0.8082381,0.0001806784,0.1874629,0.0004164162,0.00006195468,0.0002061205,0.0002647193,0.0002081856,0.002960884],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007711299,"threshold_uncertainty_score":0.01533282,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0645443691250533,"score_gpt":0.1987499315039417,"score_spread":0.1342055623788884,"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."}}