{"id":"W4288365067","doi":"10.48550/arxiv.1904.10990","title":"A Robust Approach for Securing Audio Classification Against Adversarial\\n Attacks","year":2019,"lang":"","type":"preprint","venue":"arXiv (Cornell University)","topic":"Digital Media Forensic Detection","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"École de Technologie Supérieure; Université du Québec à Montréal","funders":"","keywords":"Computer science; Spectrogram; Artificial intelligence; Codebook; Speech recognition; Audio signal; Deep learning; Smoothing; Pattern recognition (psychology); Preprocessor; Support vector machine; Feature learning; Machine learning; Speech coding; Computer vision","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":[0.0008017302,0.0009294106,0.0007080194,0.0008188109,0.0004064111,0.001074184,0.001281711,0.001289677,0.002502093],"category_scores_gemma":[0.003068833,0.0003654919,0.0007712506,0.0003923196,0.001003347,0.001408494,0.001888364,0.001754788,0.001778861],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005859095,"about_ca_system_score_gemma":0.0006323716,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007323418,"about_ca_topic_score_gemma":0.0006775848,"domain_scores_codex":[0.9986765,0.0001797794,0.00007936131,0.000278298,0.0006555306,0.0001304897],"domain_scores_gemma":[0.9988312,0.0003227468,0.000169027,0.0004070102,0.0002121227,0.00005787531],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004735833,0.0001282022,0.0008947966,0.0001947449,0.0001294854,0.000492716,0.0001311491,0.3483492,0.1899049,0.05115349,0.006642481,0.4015053],"study_design_scores_gemma":[0.00001014687,0.000143299,0.0003334327,0.00002000505,0.00001966699,0.0002349044,0.00001780311,0.9604412,0.02578237,0.007848658,0.005124742,0.00002371308],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.008824686,0.0003625291,0.98699,0.0001731197,0.0001032265,0.0000521279,0.00005459144,0.001368524,0.002071187],"genre_scores_gemma":[0.6657925,0.0009845411,0.322136,0.0004372292,0.0003661806,0.0001883448,0.000310447,0.0001775949,0.0096072],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002502093,"threshold_uncertainty_score":0.00837028,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1417010169164836,"score_gpt":0.1900639026976706,"score_spread":0.04836288578118697,"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."}}