{"id":"W4407829771","doi":"10.1109/mdat.2025.3544537","title":"Understanding Error Propagation in Deep-Learning Neural Networks’ Accelerators and Applications","year":2025,"lang":"en","type":"article","venue":"IEEE Design and Test","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Deep learning; Artificial neural network; Artificial intelligence; Backpropagation; Deep neural networks; Machine learning","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":[0.001605681,0.0006845555,0.0003768649,0.00079009,0.0003144023,0.001257021,0.0008426644,0.001049655,0.002650448],"category_scores_gemma":[0.009664254,0.0005287502,0.0003113232,0.0005821566,0.0006617602,0.002411394,0.0005126802,0.002478258,0.0005560099],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009350312,"about_ca_system_score_gemma":0.0006688687,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002840776,"about_ca_topic_score_gemma":0.003790613,"domain_scores_codex":[0.9994211,0.00008209598,0.00003166214,0.00007104963,0.0003486087,0.00004551519],"domain_scores_gemma":[0.996651,0.001862765,0.0001612868,0.0001930456,0.001070088,0.0000617169],"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":[0.0004653352,0.0001022065,0.006810652,0.0006432658,0.0001544191,0.0004989058,0.0003814708,0.22977,0.032279,0.1145874,0.07218299,0.5421243],"study_design_scores_gemma":[0.00003390365,0.0001270705,0.002396641,0.0001258605,0.0000452133,0.0003145279,0.00005992737,0.8126553,0.03639842,0.09503151,0.05276332,0.00004828801],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06296156,0.01558426,0.8732055,0.02091956,0.006769149,0.00006926899,0.000504351,0.003461394,0.01652503],"genre_scores_gemma":[0.7045414,0.0128843,0.2284975,0.004536161,0.003433686,0.00007518157,0.0007149571,0.001317536,0.04399918],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002840776,"threshold_uncertainty_score":0.008866608,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06211950306999332,"score_gpt":0.2695032183821018,"score_spread":0.2073837153121085,"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."}}