{"id":"W2951458036","doi":"10.1007/s12652-019-01357-4","title":"A hybrid GPU-FPGA based design methodology for enhancing machine learning applications performance","year":2019,"lang":"en","type":"article","venue":"Journal of Ambient Intelligence and Humanized Computing","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":29,"is_retracted":false,"has_abstract":false,"ca_institutions":"Ericsson (Canada); Université du Québec à Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Computer science; Bottleneck; Field-programmable gate array; Leverage (statistics); Graphics processing unit; Artificial neural network; Convolutional neural network; Computer engineering; CUDA; General-purpose computing on graphics processing units; Gate array; Deep learning; Exploit; Graphics; Artificial intelligence; Machine learning; Computer architecture; Embedded system; Parallel computing; Operating system","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001296318,0.0004631699,0.0002378413,0.0004251356,0.0002503197,0.000540354,0.001073018,0.0003029024,0.004791007],"category_scores_gemma":[0.0003140129,0.0001728214,0.0002319326,0.0003694344,0.0001228846,0.0004266182,0.0002360494,0.0003672431,0.00110143],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003658801,"about_ca_system_score_gemma":0.0005766241,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001184377,"about_ca_topic_score_gemma":0.003231351,"domain_scores_codex":[0.9998575,0.00002280536,0.000007910944,0.00002226454,0.00006277005,0.00002664251],"domain_scores_gemma":[0.9998167,0.00002629559,0.00001857145,0.00003287293,0.00009149659,0.00001396061],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005714643,0.0002615059,0.002546012,0.0005726145,0.0001534282,0.0004698841,0.0001187418,0.08767021,0.4612119,0.01629756,0.0160441,0.4140825],"study_design_scores_gemma":[0.00010027,0.001093713,0.002250264,0.00005668697,0.0001055636,0.0008694638,0.00004864325,0.7441606,0.2007617,0.003629846,0.04688286,0.00004037077],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1133648,0.001342658,0.8519815,0.0003658439,0.0004561734,0.0001700643,0.0002012825,0.004563889,0.0275538],"genre_scores_gemma":[0.6379609,0.0004006607,0.3460475,0.0002147226,0.00006947383,0.0001098397,0.0002148882,0.0002624687,0.01471948],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004791007,"threshold_uncertainty_score":0.01602757,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08211388848623831,"score_gpt":0.320849648271659,"score_spread":0.2387357597854207,"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."}}