{"id":"W3196680848","doi":"10.1016/j.jpdc.2021.08.003","title":"Energy efficient spiking neural network processing using approximate arithmetic units and variable precision weights","year":2021,"lang":"en","type":"article","venue":"Journal of Parallel and Distributed Computing","topic":"Advanced Memory and Neural Computing","field":"Engineering","cited_by":16,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Saskatchewan","funders":"Institute for Information and Communications Technology Promotion; Ministry of Science and ICT, South Korea","keywords":"Computer science; Variable (mathematics); Field-programmable gate array; Neuromorphic engineering; Double-precision floating-point format; Artificial neural network; Algorithm; Arbitrary-precision arithmetic; Energy (signal processing); Computation; Computer hardware; Efficient energy use; Parallel computing; Artificial intelligence; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002888459,0.0002064766,0.0003703219,0.0000663004,0.0003649517,0.0001526185,0.00008831122,0.00007163955,0.000001784956],"category_scores_gemma":[0.00005740745,0.0001854869,0.00004207889,0.000539595,0.00003298323,0.0001640783,0.0001308318,0.0003182908,5.522831e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004285258,"about_ca_system_score_gemma":0.00004671634,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001621661,"about_ca_topic_score_gemma":3.805096e-7,"domain_scores_codex":[0.99863,0.00007043996,0.0005412133,0.0001907683,0.0001873308,0.0003802116],"domain_scores_gemma":[0.9991185,0.0001744308,0.0002441273,0.00008271571,0.0002245463,0.000155653],"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":[0.00002048317,0.00001469505,0.000222916,0.00009905107,0.00002334263,0.000119072,0.00008451657,0.9766819,0.004122246,0.0002560547,0.00001020164,0.01834554],"study_design_scores_gemma":[0.0005557188,0.00003575414,0.0003293846,0.0006132383,0.00004936977,0.001414048,0.0001161224,0.9944009,0.001102383,0.001021244,0.0001579012,0.0002039296],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4777273,0.004379757,0.5176066,0.000008524409,0.00019676,0.00002194809,0.000001711402,0.00003161023,0.00002583669],"genre_scores_gemma":[0.9580295,0.00007302976,0.04146815,0.00002834064,0.0003705103,2.002129e-7,0.000007292944,0.00001989537,0.00000309936],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.4803022,"threshold_uncertainty_score":0.7563938,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01809231094196755,"score_gpt":0.232126785407456,"score_spread":0.2140344744654885,"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."}}