{"id":"W3167468823","doi":"10.3390/appliedmath2020011","title":"Gradient-Free Neural Network Training via Synaptic-Level Reinforcement Learning","year":2022,"lang":"en","type":"article","venue":"AppliedMath","topic":"Advanced Memory and Neural Computing","field":"Engineering","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Rehabilitation Institute; University of Toronto; University Health Network","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs","keywords":"Computer science; Artificial intelligence; Artificial neural network; Connectionism; Reinforcement learning; Synaptic weight; Gradient descent; Machine learning; Competitive learning; Types of artificial neural networks; Recurrent neural network; Time delay neural network","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":[],"consensus_categories":[],"category_scores_codex":[0.0002164373,0.000195351,0.0002019155,0.00004620798,0.0005742416,0.00001946336,0.0003229722,0.00002705237,0.0001118954],"category_scores_gemma":[0.00001214486,0.0002218604,0.00006533271,0.0002393772,0.00002017131,0.00008039062,0.0003178273,0.0005996157,0.00001718845],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008939418,"about_ca_system_score_gemma":0.000007105451,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001742512,"about_ca_topic_score_gemma":0.000001170224,"domain_scores_codex":[0.9987195,0.00002586556,0.0002598505,0.0002228968,0.000225741,0.0005461266],"domain_scores_gemma":[0.9994853,0.00008981166,0.00005608115,0.0002738453,0.000007592862,0.00008740199],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001324853,0.000004273666,0.00002116178,0.0000184461,0.00002714787,0.00002328899,0.0007620892,0.9779994,0.002745929,0.002359255,0.000404797,0.015621],"study_design_scores_gemma":[0.0007119628,0.0001394573,0.000116205,0.00001347848,0.00002379198,0.0001047617,0.0007547154,0.9781577,0.0008063124,0.002503258,0.01619478,0.0004735825],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7909077,0.0004610064,0.1920025,0.00007355204,0.001497748,0.0005102892,0.000004671424,0.001735572,0.01280698],"genre_scores_gemma":[0.9973395,0.000006806274,0.00185401,0.0001683274,0.000263687,0.00008136827,0.00002030673,0.00005470298,0.0002112177],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2064319,"threshold_uncertainty_score":0.9047203,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02981287971162394,"score_gpt":0.2155646530204878,"score_spread":0.1857517733088638,"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."}}