{"id":"W2905515056","doi":"10.1109/iiswc.2018.8573527","title":"Memory Requirements for Convolutional Neural Network Hardware Accelerators","year":2018,"lang":"en","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":73,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Dataflow; Bandwidth (computing); Memory bandwidth; Computer architecture; Convolutional neural network; High memory; Memory management; Deep learning; Embedded system; Computer hardware; Parallel computing; Artificial intelligence; Semiconductor memory; Computer network","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":[0.0007985043,0.0007497815,0.0004418045,0.0008618384,0.0005926927,0.001375604,0.002131799,0.0006181381,0.01347579],"category_scores_gemma":[0.004308211,0.0003914974,0.0003820759,0.00106725,0.0002869091,0.002644471,0.0008167408,0.0006056152,0.00137273],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001590049,"about_ca_system_score_gemma":0.00143134,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002887474,"about_ca_topic_score_gemma":0.004983038,"domain_scores_codex":[0.9993052,0.0001005487,0.00007309294,0.0001017538,0.000193054,0.000226191],"domain_scores_gemma":[0.9981465,0.0009541844,0.0001686353,0.0001844084,0.0004488941,0.00009733445],"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":[0.004094111,0.0003697004,0.01063176,0.001623601,0.0002813081,0.0007915167,0.0003649677,0.358606,0.1456713,0.08328229,0.03962986,0.3546537],"study_design_scores_gemma":[0.0001374993,0.0007581937,0.002507025,0.0001167435,0.0001665157,0.000404783,0.0002245617,0.8612697,0.09684001,0.01990651,0.01760881,0.0000595984],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7316949,0.005194244,0.2063008,0.002882516,0.0005445686,0.0002152198,0.0020364,0.006493928,0.04463753],"genre_scores_gemma":[0.9569796,0.00071748,0.03594568,0.000291399,0.00004330277,0.00009953346,0.0008281888,0.0001820509,0.004912704],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01347579,"threshold_uncertainty_score":0.04508108,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05564697386767817,"score_gpt":0.314030412646272,"score_spread":0.2583834387785938,"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."}}