{"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/fortified-networks-improving-the-robustness","title":"Fortified Networks: Improving the Robustness of Deep Networks by Modeling the Manifold of Hidden Representations","arxiv_id":"1804.02485","date":"2018-04-07","proceeding":"ICLR 2019 5","authors":["Alex Lamb","Jonathan Binas","Anirudh Goyal","Dmitriy Serdyuk","Sandeep Subramanian","Ioannis Mitliagkas","Yoshua Bengio"],"abstract":"Deep networks have achieved impressive results across a variety of important\ntasks. However a known weakness is a failure to perform well when evaluated on\ndata which differ from the training distribution, even if these differences are\nvery small, as is the case with adversarial examples. We propose Fortified\nNetworks, a simple transformation of existing networks, which fortifies the\nhidden layers in a deep network by identifying when the hidden states are off\nof the data manifold, and maps these hidden states back to parts of the data\nmanifold where the network performs well. Our principal contribution is to show\nthat fortifying these hidden states improves the robustness of deep networks\nand our experiments (i) demonstrate improved robustness to standard adversarial\nattacks in both black-box and white-box threat models; (ii) suggest that our\nimprovements are not primarily due to the gradient masking problem and (iii)\nshow the advantage of doing this fortification in the hidden layers instead of\nthe input space.","url_abs":"http://arxiv.org/abs/1804.02485v1","url_pdf":"http://arxiv.org/pdf/1804.02485v1.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":"fortified-networks-improving-the-robustness","repo_url":"https://github.com/jbinas/fortified-networks","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}