{"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/adversarial-attacks-on-neural-network","title":"Adversarial Attacks on Neural Network Policies","arxiv_id":"1702.02284","date":"2017-02-08","proceeding":null,"authors":["Sandy Huang","Nicolas Papernot","Ian Goodfellow","Yan Duan","Pieter Abbeel"],"abstract":"Machine learning classifiers are known to be vulnerable to inputs maliciously\nconstructed by adversaries to force misclassification. Such adversarial\nexamples have been extensively studied in the context of computer vision\napplications. In this work, we show adversarial attacks are also effective when\ntargeting neural network policies in reinforcement learning. Specifically, we\nshow existing adversarial example crafting techniques can be used to\nsignificantly degrade test-time performance of trained policies. Our threat\nmodel considers adversaries capable of introducing small perturbations to the\nraw input of the policy. We characterize the degree of vulnerability across\ntasks and training algorithms, for a subclass of adversarial-example attacks in\nwhite-box and black-box settings. Regardless of the learned task or training\nalgorithm, we observe a significant drop in performance, even with small\nadversarial perturbations that do not interfere with human perception. Videos\nare available at http://rll.berkeley.edu/adversarial.","url_abs":"http://arxiv.org/abs/1702.02284v1","url_pdf":"http://arxiv.org/pdf/1702.02284v1.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":"adversarial-attacks-on-neural-network","repo_url":"https://github.com/ssg-research/ad3-action-distribution-divergence-detector","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1702.02284","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}