{"id":"W3115877967","doi":"10.1109/msp.2020.2982820","title":"Deep Neural Network Perception Models and Robust Autonomous Driving Systems: Practical Solutions for Mitigation and Improvement","year":2020,"lang":"en","type":"article","venue":"IEEE Signal Processing Magazine","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Perception; Computer science; Human error; Safe driving; Risk analysis (engineering); Advanced driver assistance systems; Automotive engineering; Simulation; Transport engineering; Computer security; Artificial intelligence; Engineering; Business; Psychology","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":[0.00135636,0.001439176,0.0007414593,0.0004239749,0.0002895338,0.001025955,0.001375135,0.001969645,0.002518464],"category_scores_gemma":[0.004271253,0.0006429624,0.000456841,0.0005194907,0.001036583,0.002012413,0.001726013,0.003033272,0.0005956219],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009407072,"about_ca_system_score_gemma":0.0009047501,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006376546,"about_ca_topic_score_gemma":0.004334919,"domain_scores_codex":[0.9996054,0.0001032737,0.00002413721,0.0001054974,0.0001123218,0.00004932526],"domain_scores_gemma":[0.9989344,0.0005623181,0.0001262005,0.0001009724,0.0002348491,0.00004122459],"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":[0.000105656,0.00007374288,0.0009170691,0.0001852305,0.00008668385,0.00005345624,0.00007845496,0.7271429,0.003490651,0.02554706,0.006636299,0.2356828],"study_design_scores_gemma":[0.000003832884,0.00002260377,0.0001374171,0.00001333835,0.000007347143,0.000009917408,0.000009671093,0.987553,0.0005263692,0.01059986,0.001109961,0.000006699105],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007989878,0.005598645,0.9803335,0.002840742,0.0001855204,0.00002232771,0.00009186309,0.0004829853,0.002454568],"genre_scores_gemma":[0.7781361,0.008568833,0.2033892,0.0009278742,0.0006413075,0.0001371719,0.0003764882,0.0001811349,0.007641768],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006376546,"threshold_uncertainty_score":0.01267886,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04065619632948921,"score_gpt":0.2747858463175401,"score_spread":0.2341296499880508,"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."}}