{"id":"W2883283076","doi":"10.1109/isca.2018.00070","title":"Gist: Efficient Data Encoding for Deep Neural Network Training","year":2018,"lang":"en","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":146,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Microsoft Research; National Science Foundation","keywords":"Computer science; Memory footprint; Feature (linguistics); Encoding (memory); Deep learning; Bottleneck; Artificial neural network; Artificial intelligence; Embedded 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":[0.0002753734,0.0001201603,0.0001217385,0.00002879995,0.0004349051,0.0001060373,0.002016164,0.00003153798,0.00001380342],"category_scores_gemma":[0.00005354388,0.0001074497,0.00003288367,0.0004780306,0.00008305786,0.0003083551,0.0007700147,0.00007686591,0.00003626948],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001966354,"about_ca_system_score_gemma":0.0000203258,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002032601,"about_ca_topic_score_gemma":0.00002846478,"domain_scores_codex":[0.9985248,0.00001983316,0.0002137385,0.0006178697,0.000141762,0.0004820222],"domain_scores_gemma":[0.9980487,0.000311119,0.00008014904,0.001389914,0.00006896533,0.0001011017],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001293472,0.00004695776,0.00006690504,0.00000764189,0.00001651354,0.000002690532,0.0008692012,0.1360841,0.00083931,0.4221443,0.01417752,0.4257319],"study_design_scores_gemma":[0.0001378188,0.00004554278,0.00007353712,0.000004346975,0.000003771391,0.00001005539,0.00001880486,0.9671843,0.0001483964,0.003819992,0.02841625,0.0001371961],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001023575,0.00005833156,0.9918216,0.001224262,0.0005206084,0.0003411357,0.000003565672,0.0003266784,0.00468018],"genre_scores_gemma":[0.5771526,0.000001610515,0.4209389,0.000842842,0.0008742081,0.00003873287,0.00001314387,0.00001025135,0.00012765],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8311002,"threshold_uncertainty_score":0.438167,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1120225948612302,"score_gpt":0.3337931831143889,"score_spread":0.2217705882531587,"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."}}