{"id":"W2959756691","doi":"10.48550/arxiv.1907.04164","title":"Which Algorithmic Choices Matter at Which Batch Sizes? Insights From a Noisy Quadratic Model","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Quadratic equation; Artificial neural network; Stochastic gradient descent; Gradient descent; Computer science; Acceleration; Batch processing; Simple (philosophy); Momentum (technical analysis); Mathematical optimization; Algorithm; Mathematics; Artificial intelligence","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000136057,0.0006531458,0.0006760088,0.000515746,0.000230558,0.0003039915,0.003391873,0.0006236581,0.000141299],"category_scores_gemma":[0.00005758588,0.0007292797,0.0002021686,0.00121648,0.00008039871,0.000826233,0.004180228,0.0007556843,0.0009117454],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005845025,"about_ca_system_score_gemma":0.0003784382,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004785448,"about_ca_topic_score_gemma":0.0003903288,"domain_scores_codex":[0.9963819,0.0001781435,0.0004332221,0.002202013,0.0002861862,0.0005185486],"domain_scores_gemma":[0.9956245,0.0002777079,0.00055421,0.002677853,0.0006169167,0.0002488326],"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":[0.00002389008,0.0001200056,0.003098143,0.00008291457,0.0001391403,0.00002902087,0.0008116183,0.9685947,0.0000881029,0.02577904,0.001205428,0.00002803347],"study_design_scores_gemma":[0.0004889295,0.00004050083,0.001030671,0.000176961,0.00009276025,0.000002812088,0.00002050677,0.9406713,0.0002284361,0.05644464,0.00004395441,0.0007585526],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1127089,0.00005749972,0.8801127,0.0002110744,0.0007314209,0.0007873273,0.0000461592,0.000628569,0.004716328],"genre_scores_gemma":[0.9473842,0.00006885463,0.04822747,0.0002730886,0.00008187971,0.000009741536,0.00009937994,0.00005460961,0.003800782],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8346753,"threshold_uncertainty_score":0.9998662,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03901067719968814,"score_gpt":0.1875669450614755,"score_spread":0.1485562678617874,"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."}}