{"id":"W4220855081","doi":"10.3389/felec.2022.825077","title":"Exploiting Non-idealities of Resistive Switching Memories for Efficient Machine Learning","year":2022,"lang":"en","type":"article","venue":"Frontiers in Electronics","topic":"Advanced Memory and Neural Computing","field":"Engineering","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"Perimeter Institute; University of Waterloo; York University; Institut interdisciplinaire d'innovation technologique; Université de Sherbrooke","funders":"Fonds de recherche du Québec – Nature et technologies; CHIST-ERA; Natural Sciences and Engineering Research Council of Canada; Ministero dello Sviluppo Economico; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund; Agence Nationale de la Recherche; Canada Research Chairs; Government of Canada; Institut Périmètre de physique théorique; Innovation, Science and Economic Development Canada","keywords":"Computer science; Robustness (evolution); Overfitting; Resistive random-access memory; Artificial neural network; Memristor; Neuromorphic engineering; Implementation; Artificial intelligence; Spiking neural network; MNIST database; Inefficiency; Deep learning; Computer engineering; Computer architecture; Machine learning; Electronic engineering; Engineering; Voltage; Electrical engineering; Software engineering","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0003934701,0.0001404014,0.0002654634,0.0001537867,0.0002851346,0.000008407522,0.0001563956,0.00002590712,0.000003291743],"category_scores_gemma":[0.00009480545,0.0001683188,0.00006702561,0.0002526454,0.00001727875,0.00005550307,0.00007103181,0.0005476346,9.543023e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003452524,"about_ca_system_score_gemma":0.00003482665,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003626565,"about_ca_topic_score_gemma":0.000005580952,"domain_scores_codex":[0.9989008,0.00004629778,0.0002912445,0.0001771842,0.0001467906,0.0004377278],"domain_scores_gemma":[0.9996168,0.000152627,0.00008241892,0.0001017302,0.00002130456,0.00002513347],"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":[0.00007526309,0.00001111826,0.0004650621,0.0001089855,0.0000231879,0.000002632273,0.001567437,0.9836467,0.006707925,0.0003801127,0.0001011557,0.006910418],"study_design_scores_gemma":[0.0005454324,0.0002394023,0.0000379924,0.00003800313,0.00001481119,0.000004926988,0.004418215,0.9340548,0.05420168,0.002075409,0.004127475,0.000241878],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6563143,0.005548614,0.3367692,0.00001551684,0.0006850718,0.0002530881,0.00001197172,0.0001321755,0.0002701],"genre_scores_gemma":[0.9911579,0.00006653223,0.008495444,0.00001305811,0.00005509163,0.0000646832,0.00001719193,0.00004203709,0.00008801314],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3348437,"threshold_uncertainty_score":0.686384,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006919338303344787,"score_gpt":0.2090740498773749,"score_spread":0.2021547115740301,"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."}}