{"id":"W4225885311","doi":"10.23919/date54114.2022.9774635","title":"FitAct: Error Resilient Deep Neural Networks via Fine-Grained Post-Trainable Activation Functions","year":2022,"lang":"en","type":"preprint","venue":"2022 Design, Automation & Test in Europe Conference & Exhibition (DATE)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Computer science; Fault tolerance; Inference; Redundancy (engineering); Resilience (materials science); Activation function; Deep neural networks; Artificial neural network; Latency (audio); Word error rate; Artificial intelligence; Distributed computing","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","sts","scholarly_communication","research_integrity","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002923349,0.001223944,0.001017708,0.00153252,0.001319769,0.001319124,0.002608144,0.0006228258,0.004583409],"category_scores_gemma":[0.005290717,0.001420909,0.0003042197,0.003403094,0.0002125279,0.00201748,0.003005093,0.003729871,0.0006493711],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007648318,"about_ca_system_score_gemma":0.0008202291,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004701947,"about_ca_topic_score_gemma":0.0007342504,"domain_scores_codex":[0.9892236,0.003208442,0.00201811,0.002616378,0.001666592,0.001266882],"domain_scores_gemma":[0.9913821,0.001763577,0.001961865,0.002878846,0.001655541,0.0003580609],"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":[0.0001051324,0.0004672945,0.0004644037,0.0001526458,0.00005002105,0.00002913327,0.002146392,0.9728523,0.005140828,0.001933521,0.005525155,0.01113314],"study_design_scores_gemma":[0.001067627,0.0001839198,0.01216736,0.0002762103,0.00007511657,0.00006626532,0.0001124161,0.9539567,0.00006619617,0.001781527,0.02865667,0.001589954],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0155415,0.0001005738,0.9727319,0.003130312,0.002397553,0.002342762,0.0001046692,0.001925432,0.001725281],"genre_scores_gemma":[0.8112643,0.00007467821,0.160725,0.001149835,0.0006408004,0.001409559,0.01431306,0.0002928713,0.01012999],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.812007,"threshold_uncertainty_score":0.9999804,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04523905429616132,"score_gpt":0.2870918710591243,"score_spread":0.241852816762963,"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."}}