{"id":"W3197661641","doi":"10.1007/s11042-021-11394-x","title":"Protecting image privacy through adversarial perturbation","year":2021,"lang":"en","type":"article","venue":"Multimedia Tools and Applications","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University","funders":"","keywords":"Computer science; Adversarial system; Computer security; Image (mathematics); Privacy protection; Artificial intelligence; Computer vision; Theoretical computer science; Human–computer interaction","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":[0.002340815,0.0008535474,0.0008989418,0.0008442481,0.0006124951,0.001942121,0.001254765,0.002029294,0.001955793],"category_scores_gemma":[0.01504339,0.0004578804,0.0005961245,0.0008922559,0.003116612,0.00386251,0.003427116,0.003365746,0.000682192],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008497045,"about_ca_system_score_gemma":0.0005919907,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005498501,"about_ca_topic_score_gemma":0.0003414242,"domain_scores_codex":[0.9969091,0.001085239,0.00008925382,0.0005165653,0.001179765,0.0002201254],"domain_scores_gemma":[0.992214,0.004390042,0.0005015319,0.002476501,0.000317011,0.0001009282],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004368191,0.0001333807,0.001154714,0.000190525,0.0001526656,0.0005131739,0.0002241118,0.4541462,0.03146987,0.3772567,0.006889404,0.1274323],"study_design_scores_gemma":[0.00001143071,0.00004312697,0.0002854718,0.00002673023,0.00001801799,0.0002569465,0.00003754043,0.8009207,0.01165992,0.1840556,0.002664567,0.00001996391],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01425286,0.0004543687,0.9786481,0.001043302,0.00009003234,0.0000339848,0.0001025808,0.0003937194,0.004981078],"genre_scores_gemma":[0.8884223,0.0009411547,0.1030819,0.0006747537,0.0002610341,0.00008148204,0.0001946436,0.000157369,0.006185493],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002340815,"threshold_uncertainty_score":0.01237953,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02351848282571242,"score_gpt":0.2815788717922112,"score_spread":0.2580603889664989,"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."}}