{"id":"W2912975123","doi":"10.1109/reconfig.2018.8641735","title":"An Efficient FPGA-based Overlay Inference Architecture for Fully Connected DNNs","year":2018,"lang":"en","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Field-programmable gate array; Benchmark (surveying); Overlay; Inference; Computer architecture; Layer (electronics); Artificial neural network; Embedded system; Computer engineering; Computer hardware; Parallel computing; Artificial intelligence","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":[0.000196155,0.0005009025,0.0003144008,0.0004156041,0.0002886807,0.0004692334,0.001252877,0.0003289059,0.004439014],"category_scores_gemma":[0.0005273949,0.0002315301,0.0002154106,0.0003594,0.0002074629,0.001003603,0.0005839218,0.0004480231,0.0007584051],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007430203,"about_ca_system_score_gemma":0.0007066136,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005042722,"about_ca_topic_score_gemma":0.009549963,"domain_scores_codex":[0.9998117,0.00002012632,0.00001390501,0.00004802897,0.00007464727,0.00003158028],"domain_scores_gemma":[0.9998243,0.0000369943,0.00001485151,0.00003685774,0.00007130533,0.00001560662],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009039383,0.0002626061,0.002932824,0.0005266665,0.0001602799,0.0005955127,0.000189213,0.2175226,0.1216139,0.01366987,0.01800899,0.6236135],"study_design_scores_gemma":[0.00006526834,0.0003009335,0.001241076,0.00003079447,0.00006356984,0.0002316377,0.00004054533,0.9333982,0.0510084,0.00345206,0.0101383,0.00002904118],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1396455,0.001201724,0.8316691,0.0002637561,0.0002677087,0.0001601369,0.0004854852,0.01207799,0.01422863],"genre_scores_gemma":[0.7668731,0.0003297243,0.2261957,0.0001409441,0.00004294119,0.0000817878,0.0007902561,0.00009961033,0.005445993],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005042722,"threshold_uncertainty_score":0.01485002,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01833323259648206,"score_gpt":0.3009307790181714,"score_spread":0.2825975464216893,"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."}}