{"id":"W2541839172","doi":"10.1145/3123939.3123982","title":"Bit-pragmatic deep neural network computing","year":2017,"lang":"en","type":"preprint","venue":"","topic":"Ferroelectric and Negative Capacitance Devices","field":"Engineering","cited_by":229,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Massively parallel; Multiplication (music); Convolutional neural network; Parallel computing; Efficient energy use; Computation; Artificial neural network; Inference; Inefficiency; Computer engineering; Theoretical computer science; Algorithm; Artificial intelligence","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001802078,0.0003579868,0.0004669419,0.00006315569,0.0001733556,0.0001986445,0.0005549855,0.0002421769,0.0001062296],"category_scores_gemma":[0.00005254047,0.0003328915,0.0001408196,0.00007525298,0.00005077074,0.00008997217,0.0002278998,0.0008321694,0.00009165071],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007479596,"about_ca_system_score_gemma":0.00001499425,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001739798,"about_ca_topic_score_gemma":0.00005181243,"domain_scores_codex":[0.9987131,0.00003622045,0.0003069248,0.0002962676,0.0001683244,0.0004791105],"domain_scores_gemma":[0.9989942,0.0001439212,0.0001281879,0.0005885122,0.0000580564,0.00008711116],"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":[0.000003042436,0.000006730925,0.0005726299,0.0006318229,0.0001742169,0.00002022036,0.0003166538,0.9632363,0.00002826181,0.00103957,0.007983838,0.02598668],"study_design_scores_gemma":[0.000106351,0.000009914051,0.001485274,0.0002049988,0.00003347042,0.000005581511,0.00002706051,0.9931791,0.00007570126,0.003747295,0.0007176838,0.0004076147],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.3693203,0.0231232,0.265851,0.0003470268,0.0123205,0.00128969,0.00001758025,0.003898039,0.3238327],"genre_scores_gemma":[0.9941911,0.0001555999,0.004275461,0.0001070057,0.0009302766,0.00001503307,0.00003111025,0.00006100308,0.0002334399],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6248708,"threshold_uncertainty_score":0.9999123,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01705274082514729,"score_gpt":0.2423718864989099,"score_spread":0.2253191456737627,"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."}}