{"id":"W4390846314","doi":"10.1007/978-3-031-42478-6_21","title":"Logarithmic Floating-Point Multipliers for Efficient Neural Network Training","year":2023,"lang":"en","type":"book-chapter","venue":"","topic":"Low-power high-performance VLSI design","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Logarithm; Floating point; Computation; Artificial neural network; Computer science; Benchmark (surveying); Multiplication (music); Piecewise; Process (computing); Arithmetic; Range (aeronautics); Point (geometry); Parallel computing; Algorithm; Computational science; Artificial intelligence; Mathematics; Engineering","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":[0.0002046078,0.001173314,0.0003388854,0.0004924616,0.0002706071,0.001016233,0.001207269,0.000475926,0.04624321],"category_scores_gemma":[0.001394748,0.0003172416,0.000223252,0.001165204,0.0003330231,0.00154066,0.0007265731,0.001627774,0.01631206],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004727958,"about_ca_system_score_gemma":0.0003729979,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006793539,"about_ca_topic_score_gemma":0.002001074,"domain_scores_codex":[0.9998709,0.00001988881,0.000008697036,0.00002714385,0.00006058233,0.0000127936],"domain_scores_gemma":[0.9997995,0.00008850393,0.00001148353,0.00003509974,0.00005778696,0.000007633922],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001336415,0.00003918743,0.00009048217,0.0003865135,0.00001810957,0.00007214947,0.00004764248,0.01840408,0.01654596,0.08072302,0.05323037,0.8303089],"study_design_scores_gemma":[0.00009229547,0.0001809599,0.0004502931,0.0003910618,0.00004506366,0.000515582,0.00007724736,0.4283045,0.05957692,0.1679347,0.3423796,0.00005176991],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003307247,0.006036087,0.9237649,0.0005435927,0.0007700364,0.00005508434,0.0003789308,0.00308946,0.06205452],"genre_scores_gemma":[0.09584002,0.007172015,0.7302996,0.0006309481,0.0004742481,0.0002362961,0.001133947,0.001105723,0.1631072],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.04624321,"threshold_uncertainty_score":0.154699,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03163950479792249,"score_gpt":0.2137221368825604,"score_spread":0.1820826320846379,"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."}}