{"id":"W4287867706","doi":"10.48550/arxiv.2002.07711","title":"An Energy-Efficient Accelerator Architecture with Serial Accumulation\\n Dataflow for Deep CNNs","year":2020,"lang":"","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Dram; Computer science; Dataflow; Convolutional neural network; Computation; Convolution (computer science); Energy consumption; Parallel computing; CAS latency; Computer hardware; Latency (audio); Dataflow architecture; Embedded system; Efficient energy use; Hardware acceleration; Architecture; Energy (signal processing); Artificial neural network; Artificial intelligence; Field-programmable gate array; Algorithm; Semiconductor memory","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001810801,0.0009923107,0.0007769924,0.0002956899,0.001074514,0.0005190716,0.004795989,0.0004715342,0.0000620752],"category_scores_gemma":[0.00003691724,0.001088039,0.0003240176,0.002184989,0.0003548888,0.0008049929,0.002100766,0.0009296743,0.00004544362],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003491972,"about_ca_system_score_gemma":0.0005907166,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004405717,"about_ca_topic_score_gemma":0.000239502,"domain_scores_codex":[0.9939102,0.0002726069,0.0005526299,0.003989803,0.000272094,0.001002654],"domain_scores_gemma":[0.9939145,0.0003897349,0.0008127796,0.003392174,0.0005636194,0.0009271316],"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":[0.0005805779,0.0002758196,0.00008271597,0.00004858927,0.0001131757,0.00006403419,0.0003350356,0.8587502,0.0004960552,0.135164,0.00006788793,0.004021931],"study_design_scores_gemma":[0.001602222,0.0004620983,0.0001789881,0.00006314529,0.0002224903,0.0000125328,0.00007377615,0.9624119,0.0007041472,0.02698808,0.006087661,0.001193],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05503803,0.0000299119,0.9411312,0.0005370482,0.0006140791,0.001706521,0.0003716665,0.0004381434,0.0001334325],"genre_scores_gemma":[0.9390382,0.00007251465,0.05858266,0.0005044769,0.0009335968,0.0000310671,0.0005762902,0.0001009446,0.0001602992],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8840001,"threshold_uncertainty_score":0.999157,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.113628988488602,"score_gpt":0.2347072290643945,"score_spread":0.1210782405757925,"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."}}