{"id":"W2400062642","doi":"10.1016/j.neucom.2019.11.021","title":"Annealed gradient descent for deep learning","year":2019,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"York University","funders":"National Natural Science Foundation of China","keywords":"MNIST database; Stochastic gradient descent; Deep learning; Computer science; Artificial intelligence; Gradient descent; Convex function; Schedule; Stochastic optimization; Convergence (economics); Convex optimization; Pattern recognition (psychology); Artificial neural network; Regular polygon; Mathematical optimization; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001158722,0.0009388506,0.001369852,0.0006326742,0.0004723258,0.001008243,0.001404339,0.002037073,0.003649699],"category_scores_gemma":[0.005014408,0.0008333681,0.0005937803,0.0009889515,0.00131097,0.001460694,0.001431223,0.002708076,0.0009153865],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001443902,"about_ca_system_score_gemma":0.001499506,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007279912,"about_ca_topic_score_gemma":0.00911138,"domain_scores_codex":[0.9996302,0.0001432144,0.00002027367,0.00006223036,0.0001098431,0.00003420159],"domain_scores_gemma":[0.9988235,0.0007092055,0.00007428374,0.0001198483,0.0002050303,0.00006812006],"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":[0.00009010973,0.00006428396,0.0003148353,0.0001802266,0.00006494801,0.00005497261,0.00006067113,0.7430351,0.001684821,0.1683527,0.008262707,0.07783461],"study_design_scores_gemma":[0.000004062465,0.000006371121,0.0000310391,0.000006679457,0.000003183649,0.000004136949,0.00000153536,0.9697487,0.0001592838,0.02939849,0.0006333278,0.00000309366],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.005229451,0.001211459,0.9903273,0.0003798191,0.0001475032,0.00002436081,0.00006839428,0.0003938926,0.002217814],"genre_scores_gemma":[0.3895148,0.002154878,0.5756512,0.0005010584,0.0004517747,0.0003782546,0.0005843032,0.0008054461,0.02995834],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007279912,"threshold_uncertainty_score":0.01447505,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01079820881875083,"score_gpt":0.2311645944071059,"score_spread":0.2203663855883551,"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."}}