{"id":"W4393579178","doi":"10.1145/3626202.3637585","title":"E4SA: An Ultra-Efficient Systolic Array Architecture for 4-Bit Convolutional Neural Networks","year":2024,"lang":"en","type":"article","venue":"","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Computer science; Convolutional neural network; Parallel computing; Digital signal processing; Systolic array; Field-programmable gate array; Computer engineering; Quantization (signal processing); Overhead (engineering); Kernel (algebra); Computer hardware; Algorithm; Embedded system; Very-large-scale integration; Artificial intelligence","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":[0.0002324986,0.0004387397,0.0002347949,0.0005762302,0.0002774471,0.0004295869,0.001268067,0.0003432642,0.002803952],"category_scores_gemma":[0.0003817918,0.0002120078,0.0003286024,0.0005417094,0.0002614953,0.0007513136,0.0005210295,0.0005014653,0.000864051],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004973047,"about_ca_system_score_gemma":0.0009033655,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001823095,"about_ca_topic_score_gemma":0.004028887,"domain_scores_codex":[0.999874,0.00001691579,0.0000145314,0.00002478622,0.00004793733,0.00002184124],"domain_scores_gemma":[0.9998197,0.00003383267,0.00003113787,0.00003478242,0.00006057327,0.00002007098],"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":[0.0006520643,0.000232804,0.003222659,0.0003428952,0.0001379189,0.0003430846,0.0002092565,0.1383417,0.2317634,0.02539681,0.01337685,0.5859805],"study_design_scores_gemma":[0.00006982771,0.0007572324,0.001435966,0.00004298753,0.00005488209,0.0003092161,0.00005062136,0.8518347,0.1025092,0.0079652,0.03491788,0.00005233306],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1023537,0.00113961,0.8774369,0.0002666179,0.0001984276,0.0001225088,0.0003402722,0.007505823,0.01063609],"genre_scores_gemma":[0.5008575,0.0005080888,0.4890423,0.0002578521,0.00005827883,0.0001744163,0.0008771358,0.0001929813,0.008031328],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002803952,"threshold_uncertainty_score":0.009380162,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01553573081770237,"score_gpt":0.2578875878322113,"score_spread":0.2423518570145089,"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."}}