{"id":"W7047209072","doi":"","title":"FPRaker: Exploiting Fine-grain Sparsity to Accelerate Neural Network Training","year":2020,"lang":"en","type":"dissertation","venue":"TSpace (University of Toronto)","topic":"Magnetic confinement fusion research","field":"Physics and Astronomy","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of Toronto","keywords":"Exploit; Accumulator (cryptography); Artificial neural network; Computation; Training (meteorology); Quantization (signal processing); Training set; Compression (physics)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004038643,0.0006278552,0.0003155702,0.0003886655,0.0003240708,0.0006531441,0.001150725,0.000467372,0.006067719],"category_scores_gemma":[0.00128802,0.0002939213,0.0003369318,0.0004129278,0.0003489941,0.001212618,0.00083789,0.001120893,0.001684141],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005109252,"about_ca_system_score_gemma":0.0009874678,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002695808,"about_ca_topic_score_gemma":0.008091342,"domain_scores_codex":[0.999836,0.0000215697,0.000008836045,0.00003634117,0.0000692527,0.00002798283],"domain_scores_gemma":[0.9996067,0.0001331563,0.00003168523,0.0000989274,0.00009529428,0.00003418712],"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":[0.0005038353,0.0002324398,0.002658783,0.0002467791,0.0001606986,0.0002141362,0.0001929252,0.2221982,0.0956158,0.01601008,0.02584784,0.6361185],"study_design_scores_gemma":[0.00003748152,0.0001557611,0.0005545401,0.0000192469,0.00002000813,0.00006170876,0.00002528252,0.951932,0.03206836,0.006570553,0.008539884,0.00001524438],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09653427,0.0009312565,0.8681463,0.0005394229,0.0002994831,0.0001086589,0.0004817001,0.01662929,0.01632955],"genre_scores_gemma":[0.4265403,0.0004707829,0.5543892,0.0003694703,0.00009849924,0.0001721055,0.00144513,0.00111287,0.01540159],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006067719,"threshold_uncertainty_score":0.02029854,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03709876029966326,"score_gpt":0.2728718151855006,"score_spread":0.2357730548858373,"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."}}