{"id":"W4390116303","doi":"10.1016/j.cose.2023.103674","title":"Improving adversarial transferability through hybrid augmentation","year":2023,"lang":"en","type":"article","venue":"Computers & Security","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"Fundamental Research Funds for the Central Universities; Basic and Applied Basic Research Foundation of Guangdong Province; National Key Research and Development Program of China; National Natural Science Foundation of China","keywords":"Transferability; Computer science; Adversarial system; Domain (mathematical analysis); Focus (optics); Masking (illustration); Artificial intelligence; Diversity (politics); Scaling; Frequency domain; Image (mathematics); Machine learning; Pattern recognition (psychology); Algorithm; Computer vision; Mathematics","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":[0.0007764255,0.0002731429,0.0003114027,0.000129123,0.0004071301,0.0002461488,0.001282957,0.00008336717,0.00002176313],"category_scores_gemma":[0.0001296119,0.00029547,0.0001755935,0.0008417229,0.0001031931,0.001483381,0.0006508976,0.0004608872,0.0001363254],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001674216,"about_ca_system_score_gemma":0.0001096035,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002813495,"about_ca_topic_score_gemma":0.0000122987,"domain_scores_codex":[0.9973512,0.0003110804,0.0004080074,0.000858088,0.0005120209,0.0005596092],"domain_scores_gemma":[0.9985424,0.0003530752,0.0001279295,0.0007669613,0.0000887694,0.0001208997],"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":[0.0002660383,0.0005624485,0.004614757,0.0005994302,0.0002975781,0.0006444529,0.05409615,0.2047935,0.002378509,0.1772172,0.01899877,0.5355312],"study_design_scores_gemma":[0.001309371,0.0001002493,0.002054059,0.00002474649,0.00001721301,0.00001387376,0.00008707914,0.962886,0.0008521555,0.02916814,0.003054095,0.0004330109],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09944413,0.00002335296,0.8934047,0.001421875,0.003453024,0.0003253136,0.000006580171,0.00142921,0.0004918326],"genre_scores_gemma":[0.9520286,0.000008634998,0.04710086,0.0003524101,0.000412463,0.00001694136,0.00004025005,0.00002135866,0.00001852247],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8525844,"threshold_uncertainty_score":0.9999498,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01467202256203794,"score_gpt":0.2663804536482066,"score_spread":0.2517084310861686,"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."}}