{"id":"W2952857977","doi":"10.1109/dac.2018.8465915","title":"Loom: Exploiting Weight and Activation Precisions to Accelerate Convolutional Neural Networks","year":2018,"lang":"en","type":"article","venue":"2018 55th ACM/ESDA/IEEE Design Automation Conference (DAC)","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":88,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Convolutional neural network; Computer science; LOOM; Granularity; Inference; Energy (signal processing); Parallel computing; Layer (electronics); Deep neural networks; Artificial neural network; Algorithm; Artificial intelligence; Mathematics; Statistics; 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":[0.0004085393,0.0011444,0.0003476231,0.0006031676,0.0003367585,0.0009006354,0.002525644,0.0005080282,0.01357895],"category_scores_gemma":[0.001674355,0.0004687334,0.0004816111,0.0006719349,0.000336903,0.001896434,0.001285801,0.001401471,0.003424072],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001022497,"about_ca_system_score_gemma":0.001544591,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004804214,"about_ca_topic_score_gemma":0.01329749,"domain_scores_codex":[0.9996769,0.00003043909,0.00002102623,0.00006689619,0.0001338373,0.00007091069],"domain_scores_gemma":[0.9995479,0.0001403083,0.00005269236,0.0001145257,0.0001099708,0.00003454842],"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":[0.001576933,0.0002998664,0.004052175,0.0007887414,0.0002140622,0.0003429834,0.0002205781,0.08024512,0.1014242,0.02638336,0.09042165,0.6940304],"study_design_scores_gemma":[0.0002961462,0.0006830869,0.001766061,0.0001149817,0.0001279425,0.0002732505,0.00006594104,0.7636839,0.1472072,0.01580543,0.06987673,0.00009939248],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09672406,0.002989376,0.8098145,0.00096212,0.0006793858,0.000186656,0.001503234,0.06229792,0.02484274],"genre_scores_gemma":[0.5366524,0.0009283714,0.423454,0.0009109944,0.0001903356,0.0003746664,0.003391776,0.002315594,0.0317819],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01357895,"threshold_uncertainty_score":0.04542613,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0962652447524089,"score_gpt":0.3100672727499506,"score_spread":0.2138020279975417,"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."}}