{"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/provable-robustness-of-relu-networks-via","title":"Provable Robustness of ReLU networks via Maximization of Linear Regions","arxiv_id":"1810.07481","date":"2018-10-17","proceeding":null,"authors":["Francesco Croce","Maksym Andriushchenko","Matthias Hein"],"abstract":"It has been shown that neural network classifiers are not robust. This raises\nconcerns about their usage in safety-critical systems. We propose in this paper\na regularization scheme for ReLU networks which provably improves the\nrobustness of the classifier by maximizing the linear regions of the classifier\nas well as the distance to the decision boundary. Our techniques allow even to\nfind the minimal adversarial perturbation for a fraction of test points for\nlarge networks. In the experiments we show that our approach improves upon\nadversarial training both in terms of lower and upper bounds on the robustness\nand is comparable or better than the state-of-the-art in terms of test error\nand robustness.","url_abs":"http://arxiv.org/abs/1810.07481v2","url_pdf":"http://arxiv.org/pdf/1810.07481v2.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":"provable-robustness-of-relu-networks-via","repo_url":"https://github.com/max-andr/provable-robustness-max-linear-regions","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"provable-robustness-of-relu-networks-via","repo_url":"https://github.com/fra31/mmr-universal","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[{"method_slug":"relu","method_name":"ReLU"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1810.07481","atlas_url":"https://app.syntology.ai/?focus=1810.07481","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}