{"id":"W4321764792","doi":"10.22541/au.167727732.20256861/v1","title":"HESSPROP: Mitigating Memristive DNN Weight Mapping Errors with Hessian Backpropagation","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Advanced Memory and Neural Computing","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University; University of Toronto","funders":"","keywords":"MNIST database; Backpropagation; Hessian matrix; Maxima and minima; Artificial neural network; Gradient descent; Stochastic gradient descent; Robustness (evolution); Computer science; Memristor; Artificial intelligence; Algorithm; Convolutional neural network; Pattern recognition (psychology); Mathematics; Applied mathematics; Electronic engineering; Engineering; Mathematical analysis","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":[0.0001802917,0.0004896807,0.0004501948,0.0001822567,0.0001912503,0.00007745954,0.0002668753,0.0002298588,0.00004002984],"category_scores_gemma":[0.00003739876,0.0004218462,0.0000997781,0.0003085195,0.00005662937,0.0001742055,0.0002873476,0.001127168,0.00009350153],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001828072,"about_ca_system_score_gemma":0.00004465077,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001478423,"about_ca_topic_score_gemma":0.00003071347,"domain_scores_codex":[0.9981411,0.00004492051,0.0004481248,0.0006099367,0.0002775835,0.0004783295],"domain_scores_gemma":[0.9991165,0.0001352911,0.0001557248,0.0003961369,0.00007446339,0.000121904],"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":[0.000017349,0.00001342586,0.0005886881,0.001878491,0.0002179245,0.0001718778,0.001258377,0.972528,0.009646637,0.0003489473,0.0004687243,0.0128615],"study_design_scores_gemma":[0.0009126168,0.0001042284,0.003630158,0.007062639,0.0001386869,0.00005669041,0.002469306,0.7065926,0.2656316,0.00880552,0.001481315,0.003114689],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5431222,0.0002923956,0.4336314,0.00034641,0.002255914,0.001091166,0.00001949678,0.005539668,0.01370134],"genre_scores_gemma":[0.9723803,0.00004874254,0.02569301,0.00004513353,0.0005179011,0.00008303636,0.0001048758,0.0001800912,0.0009469003],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.429258,"threshold_uncertainty_score":0.9998233,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03380868750380305,"score_gpt":0.2401455740023423,"score_spread":0.2063368864985393,"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."}}