{"id":"W4309505072","doi":"10.1145/3572751.3572765","title":"Characterizing I/O in Machine Learning with MLPerf Storage","year":2022,"lang":"en","type":"article","venue":"ACM SIGMOD Record","topic":"Advanced Data Storage Technologies","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Focus (optics); Inference; Software; Machine learning; Computer data storage; Training set; Training (meteorology); Data access; Database; Artificial intelligence; Computer engineering; Operating system","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":[0.000303498,0.0001661452,0.0002088313,0.000274034,0.0002769139,0.0000608659,0.0022606,0.00003545777,0.00006351681],"category_scores_gemma":[0.0002332974,0.0001556783,0.00002655265,0.0008548382,0.00004315693,0.0007629736,0.002924161,0.0008144728,0.00001692425],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000185581,"about_ca_system_score_gemma":0.00004243127,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001153128,"about_ca_topic_score_gemma":0.00006550504,"domain_scores_codex":[0.9985131,0.0001065126,0.0001949622,0.0005379151,0.0002806019,0.0003668992],"domain_scores_gemma":[0.9984758,0.000149208,0.0001416564,0.001177834,0.00001730372,0.00003817587],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001183735,0.0002057347,0.04876668,0.0000283244,0.00003058917,0.00128705,0.002273113,0.006457198,0.01959151,0.005648853,0.0002984343,0.9152941],"study_design_scores_gemma":[0.004925001,0.005212383,0.02583426,0.0001641588,0.00002511443,0.0009254786,0.002896303,0.3193644,0.01059416,0.021403,0.6050707,0.003585052],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7306154,0.0003285848,0.2640077,0.002436004,0.0004486178,0.0003032401,0.00001459122,0.001414603,0.000431354],"genre_scores_gemma":[0.8922457,0.00003116179,0.1070344,0.0002293044,0.0000202573,0.0001030446,0.00001913596,0.00002211019,0.0002949711],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9117091,"threshold_uncertainty_score":0.6348374,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01743802220997948,"score_gpt":0.2339386751193323,"score_spread":0.2165006529093528,"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."}}