{"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/training-for-faster-adversarial-robustness","title":"Training for Faster Adversarial Robustness Verification via Inducing ReLU Stability","arxiv_id":"1809.03008","date":"2018-09-09","proceeding":"ICLR 2019 5","authors":["Kai Y. Xiao","Vincent Tjeng","Nur Muhammad Shafiullah","Aleksander Madry"],"abstract":"We explore the concept of co-design in the context of neural network\nverification. Specifically, we aim to train deep neural networks that not only\nare robust to adversarial perturbations but also whose robustness can be\nverified more easily. To this end, we identify two properties of network models\n- weight sparsity and so-called ReLU stability - that turn out to significantly\nimpact the complexity of the corresponding verification task. We demonstrate\nthat improving weight sparsity alone already enables us to turn computationally\nintractable verification problems into tractable ones. Then, improving ReLU\nstability leads to an additional 4-13x speedup in verification times. An\nimportant feature of our methodology is its \"universality,\" in the sense that\nit can be used with a broad range of training procedures and verification\napproaches.","url_abs":"http://arxiv.org/abs/1809.03008v3","url_pdf":"http://arxiv.org/pdf/1809.03008v3.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":"training-for-faster-adversarial-robustness","repo_url":"https://github.com/MadryLab/relu_stable","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"adversarial-robustness","task_name":"Adversarial Robustness"}],"methods":[{"method_slug":"relu","method_name":"ReLU"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1809.03008","atlas_url":"https://app.syntology.ai/?focus=1809.03008","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}