{"id":"W4385290324","doi":"10.1109/iscas46773.2023.10181445","title":"HESSPROP: Mitigating Memristive DNN Weight Mapping Errors with Hessian Backpropagation","year":2023,"lang":"en","type":"article","venue":"","topic":"Advanced Memory and Neural Computing","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University; University of Toronto","funders":"","keywords":"Backpropagation; Hessian matrix; Computer science; Artificial intelligence; Artificial neural network; Mathematics; Applied mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.0001018208,0.0001640199,0.0001466793,0.0001003863,0.0001699397,0.00002451274,0.00008679334,0.00004224187,0.0000392625],"category_scores_gemma":[0.00001855738,0.0001328675,0.00003017672,0.0005843948,0.00002975343,0.0002417614,0.00002969804,0.0001920709,0.0001324471],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004820178,"about_ca_system_score_gemma":0.000009375236,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002252139,"about_ca_topic_score_gemma":0.000006742633,"domain_scores_codex":[0.9991264,0.00001798977,0.0001828139,0.0002104198,0.000141918,0.0003204324],"domain_scores_gemma":[0.9996549,0.0000793031,0.00003558412,0.0001311971,0.00002818014,0.00007081701],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000035663,0.00002206407,0.003751152,0.0007403778,0.0001570044,0.0003199518,0.003511937,0.6306857,0.2690246,0.002349646,0.001775441,0.08762643],"study_design_scores_gemma":[0.0006697627,0.0000805102,0.005892952,0.0005402677,0.00001968475,0.00003791009,0.002660956,0.4688852,0.5165533,0.00120451,0.002632202,0.0008227819],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9219416,0.00004310554,0.0611606,0.0001573391,0.0002643937,0.0002080968,0.00000146908,0.002163649,0.01405974],"genre_scores_gemma":[0.9933556,0.00001045729,0.005833162,0.00004377057,0.0001449194,0.0000156378,0.00001527314,0.00004316536,0.0005379828],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2475286,"threshold_uncertainty_score":0.541818,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01602854403935014,"score_gpt":0.2162758300549623,"score_spread":0.2002472860156122,"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."}}