{"id":"W4295836371","doi":"10.48550/arxiv.2105.14564","title":"Evaluating Resilience of Encrypted Traffic Classification Against Adversarial Evasion Attacks","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Internet Traffic Analysis and Secure E-voting","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Artificial intelligence; Deep learning; Convolutional neural network; Adversarial machine learning; Adversarial system; Machine learning; Evasion (ethics); Encryption; Decision tree; Traffic classification; Resilience (materials science); The Internet; Computer security","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009266448,0.0003419257,0.0005378291,0.0003315949,0.0001784661,0.0001399483,0.001984795,0.0003840608,0.00003852549],"category_scores_gemma":[0.0002219666,0.0003904148,0.0004561115,0.0009825783,0.0001632335,0.0004224998,0.001251369,0.0006800074,0.00001912321],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002671131,"about_ca_system_score_gemma":0.0005069628,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002844131,"about_ca_topic_score_gemma":0.00007236358,"domain_scores_codex":[0.9967579,0.0004889207,0.0005647968,0.001487322,0.0003361911,0.0003649247],"domain_scores_gemma":[0.9972212,0.0002074539,0.0008116977,0.0009744623,0.0006550521,0.0001301264],"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":[0.00002597592,0.0001110776,0.0001170189,0.00006848553,0.0001032105,0.00006661587,0.001199035,0.9505837,0.0008612885,0.04197344,0.00004054068,0.004849648],"study_design_scores_gemma":[0.0004317419,0.00006670196,0.0005222876,0.0003070339,0.0001228229,0.000001583918,0.0008047122,0.9970585,0.0002559689,0.00005448137,0.00001854374,0.0003556852],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5915478,0.00004856347,0.4072104,0.00002576748,0.0004474079,0.0001493452,0.000002748578,0.00009401057,0.0004739294],"genre_scores_gemma":[0.9950128,0.00009594857,0.004452877,0.00003523278,0.0000976937,9.25455e-7,0.0000592765,0.00001547428,0.0002297721],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4034649,"threshold_uncertainty_score":0.9998548,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1153685052425876,"score_gpt":0.2498503531899524,"score_spread":0.1344818479473648,"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."}}