{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learning-universal-adversarial-perturbations","title":"Learning Universal Adversarial Perturbations with Generative Models","arxiv_id":"1708.05207","date":"2017-08-17","proceeding":null,"authors":["Jamie Hayes","George Danezis"],"abstract":"Neural networks are known to be vulnerable to adversarial examples, inputs\nthat have been intentionally perturbed to remain visually similar to the source\ninput, but cause a misclassification. It was recently shown that given a\ndataset and classifier, there exists so called universal adversarial\nperturbations, a single perturbation that causes a misclassification when\napplied to any input. In this work, we introduce universal adversarial\nnetworks, a generative network that is capable of fooling a target classifier\nwhen it's generated output is added to a clean sample from a dataset. We show\nthat this technique improves on known universal adversarial attacks.","url_abs":"http://arxiv.org/abs/1708.05207v3","url_pdf":"http://arxiv.org/pdf/1708.05207v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learning-universal-adversarial-perturbations","repo_url":"https://github.com/jhayes14/UAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"graph-classification","task_name":"Graph Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/graph-classification-on-nci1","task":"Graph Classification","dataset":"NCI1","model":"DUGNN","rank_in_archive_order":10,"of":69,"metrics":{"Accuracy":"85.50%"},"uses_additional_data":true}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}