{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002499851,0.0001480351,0.0001651097,0.0001392418,0.0001761125,0.0001343573,0.0005979,0.00004253602,0.000005450986],"category_scores_gemma":[0.0001127599,0.0001534566,0.00008530615,0.0003149254,0.00001336036,0.0001755263,0.0002506952,0.0001538208,0.00003360921],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004328804,"about_ca_system_score_gemma":0.00001750563,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001879418,"about_ca_topic_score_gemma":2.024766e-7,"domain_scores_codex":[0.998616,0.00004997263,0.0002534497,0.0004971233,0.0001970445,0.0003863923],"domain_scores_gemma":[0.9990625,0.000260963,0.0001502537,0.00032782,0.0001161568,0.00008233835],"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":[0.00003123074,0.0002436566,0.01868405,0.0002070661,0.00004021361,0.00001656249,0.003066946,0.4254333,0.005207361,0.2548231,0.0006044804,0.291642],"study_design_scores_gemma":[0.0003713901,0.0002318631,0.001406039,0.00002860441,0.000003077484,0.00001667839,0.00001293778,0.9935584,0.001315194,0.0009394316,0.001946415,0.0001699514],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04506981,0.00002796854,0.9519789,0.0001552071,0.0005594048,0.0006847915,1.946466e-7,0.0008163169,0.0007074593],"genre_scores_gemma":[0.7211416,0.000001771269,0.2783895,0.0002792928,0.00005650139,0.00002527241,0.000002239449,0.00001860462,0.0000852275],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6760718,"threshold_uncertainty_score":0.625778,"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."}}