{"id":"W2997288154","doi":"10.1109/robio49542.2019.8961780","title":"Deep Learning Training with Simulated Approximate Multipliers","year":2019,"lang":"en","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"Killam Trusts; Dalhousie University","keywords":"Computer science; Convolutional neural network; Approximation error; Artificial neural network; Lagrange multiplier; Speedup; Mathematical optimization; Training (meteorology); Power (physics); Algorithm; Artificial intelligence; Mathematics; Parallel computing","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":[],"consensus_categories":[],"category_scores_codex":[0.00006776279,0.0001102538,0.0001142914,0.00004415375,0.0001113657,0.00005796472,0.0004122159,0.00003061013,0.00002922162],"category_scores_gemma":[0.000008257484,0.00008591192,0.00002229136,0.0004554982,0.00002394082,0.0003576067,0.0001014478,0.0001740959,0.000174194],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001910848,"about_ca_system_score_gemma":0.00001315023,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002199878,"about_ca_topic_score_gemma":0.000002312292,"domain_scores_codex":[0.9990693,0.00002049721,0.0001148979,0.0003689663,0.0001431032,0.0002832277],"domain_scores_gemma":[0.9993195,0.0001256585,0.00006286214,0.0003798711,0.00003717139,0.00007495986],"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":[0.000005149445,0.000009913978,0.0008971504,0.000003096302,0.000007723338,0.000002201379,0.0006960844,0.9329831,0.001194089,0.02028902,0.000004846495,0.0439076],"study_design_scores_gemma":[0.0003368599,0.0000527846,0.0003287237,0.0000063439,0.000001607594,0.000008217941,0.0001277998,0.9965107,0.0003306688,0.0006230778,0.001525085,0.0001481194],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05397997,0.00001254098,0.936061,0.0001685599,0.00003556722,0.000255534,4.856809e-8,0.0005942082,0.008892575],"genre_scores_gemma":[0.821761,0.000002272109,0.1770694,0.0001659039,0.00001160491,0.000008718725,0.00000179111,0.00001167374,0.0009676763],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.767781,"threshold_uncertainty_score":0.3503386,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01586306978373895,"score_gpt":0.240508102606348,"score_spread":0.224645032822609,"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."}}