{"id":"W2980235527","doi":"10.1109/tvlsi.2019.2940943","title":"Improving the Accuracy and Hardware Efficiency of Neural Networks Using Approximate Multipliers","year":2019,"lang":"en","type":"article","venue":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","topic":"Low-power high-performance VLSI design","field":"Engineering","cited_by":147,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"MNIST database; Computer science; Artificial neural network; Convolutional neural network; Noise (video); Multiplication (music); Artificial intelligence; Efficient energy use; Computer engineering; Deep neural networks; Machine learning; Algorithm; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0007419081,0.0006294337,0.0003849818,0.0004373832,0.0001962734,0.0007570859,0.0006143773,0.0004262355,0.001309431],"category_scores_gemma":[0.003915728,0.0002265266,0.0002490275,0.0004630642,0.0003350137,0.001068939,0.000388831,0.0005961984,0.0003729929],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004714992,"about_ca_system_score_gemma":0.0005194895,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001421924,"about_ca_topic_score_gemma":0.00190157,"domain_scores_codex":[0.999501,0.0001021028,0.00003960258,0.00008421329,0.0002184837,0.00005460363],"domain_scores_gemma":[0.9990098,0.0004850173,0.0001415471,0.0001458929,0.0002029059,0.00001473834],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003673576,0.00008635664,0.003602652,0.000306313,0.00007526343,0.0001450302,0.00007230508,0.5915998,0.05351936,0.01225437,0.001317664,0.3366536],"study_design_scores_gemma":[0.000016245,0.0001853894,0.0008760665,0.0000239328,0.00002335544,0.00007603942,0.00001726393,0.9720492,0.02184016,0.002890647,0.00199216,0.000009405941],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.283003,0.001936209,0.7033298,0.0004618795,0.000191936,0.00005403818,0.0001227112,0.001563801,0.009336649],"genre_scores_gemma":[0.8188354,0.0005153235,0.1788036,0.00006733969,0.00002969632,0.00004466227,0.00009530757,0.00005797514,0.001550743],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.001421924,"threshold_uncertainty_score":0.004380465,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009042114404894398,"score_gpt":0.2074460499332867,"score_spread":0.1984039355283924,"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."}}