{"id":"W4388455116","doi":"10.1145/3631610","title":"HyBNN: Quantifying and Optimizing Hardware Efficiency of Binary Neural Networks","year":2023,"lang":"en","type":"article","venue":"ACM Transactions on Reconfigurable Technology and Systems","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"Fundamental Research Funds for the Central Universities; National Natural Science Foundation of China; Xidian University; China Association for Science and Technology","keywords":"Computer science; Floating point; Quantization (signal processing); Overhead (engineering); Computer hardware; Binary number; Hardware acceleration; Deep learning; Artificial neural network; Computer engineering; Inference; Edge device; Algorithm; Artificial intelligence; Field-programmable gate array; Arithmetic; Mathematics","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":[0.0009484609,0.001015621,0.0003819979,0.0006149869,0.000344047,0.0007885637,0.001641096,0.0004692486,0.003766237],"category_scores_gemma":[0.003562682,0.0002772448,0.0003233551,0.0005572216,0.0004424502,0.002170123,0.0007761987,0.000756007,0.0009280987],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001077156,"about_ca_system_score_gemma":0.001569911,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005173697,"about_ca_topic_score_gemma":0.01137671,"domain_scores_codex":[0.9993708,0.000106169,0.00005164729,0.0001158514,0.0002214276,0.0001341145],"domain_scores_gemma":[0.9992079,0.0003144478,0.00007361459,0.0001550746,0.0002116207,0.0000372983],"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":[0.0007859555,0.0001963032,0.007448083,0.0006691975,0.0001359184,0.0001837663,0.0001010073,0.5087782,0.04354511,0.01534959,0.01447176,0.4083351],"study_design_scores_gemma":[0.00005017637,0.0002706845,0.001225089,0.0000459915,0.00004943503,0.0001115202,0.00004079244,0.9511194,0.03534502,0.004796183,0.006925567,0.00002010728],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4323181,0.005792606,0.5148863,0.001133704,0.0004972697,0.0002619691,0.001220235,0.01141231,0.0324775],"genre_scores_gemma":[0.8087626,0.0007372733,0.1838362,0.000333749,0.00003811109,0.0001835678,0.001097034,0.0005589323,0.004452556],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005173697,"threshold_uncertainty_score":0.01259935,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03841834910659853,"score_gpt":0.2736909359617335,"score_spread":0.235272586855135,"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."}}