{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002021411,0.001288331,0.001025376,0.002554645,0.001332804,0.002762787,0.003105329,0.001020149,0.007377703],"category_scores_gemma":[0.02144969,0.0006063778,0.0005541351,0.003338635,0.001561337,0.009864399,0.001944495,0.001341104,0.002185463],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002506387,"about_ca_system_score_gemma":0.00281151,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004600313,"about_ca_topic_score_gemma":0.003972217,"domain_scores_codex":[0.9969953,0.0002796561,0.0002357452,0.000617286,0.001093907,0.0007780642],"domain_scores_gemma":[0.9870204,0.006247272,0.000933396,0.003291525,0.002125251,0.000382211],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.003626578,0.0009542507,0.03767353,0.001108131,0.0001643953,0.001134122,0.000954856,0.2757499,0.05095065,0.08170464,0.04517791,0.500801],"study_design_scores_gemma":[0.0001034206,0.0005214356,0.008130414,0.0001553831,0.00008726779,0.0006186168,0.0003975362,0.8534338,0.04457492,0.07404394,0.01785126,0.00008193642],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6636809,0.01051381,0.276221,0.003642972,0.0007888303,0.000406122,0.004918871,0.01290465,0.02692297],"genre_scores_gemma":[0.9519822,0.001404756,0.03758749,0.0003329656,0.0002455965,0.0003001923,0.0021333,0.0008043721,0.005209235],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007377703,"threshold_uncertainty_score":0.02468091,"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."}}