{"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-noise-layer-regularize-neural","title":"Adversarial Noise Layer: Regularize Neural Network By Adding Noise","arxiv_id":"1805.08000","date":"2018-05-21","proceeding":null,"authors":["Zhonghui You","Jinmian Ye","Kunming Li","Zenglin Xu","Ping Wang"],"abstract":"In this paper, we introduce a novel regularization method called Adversarial\nNoise Layer (ANL) and its efficient version called Class Adversarial Noise\nLayer (CANL), which are able to significantly improve CNN's generalization\nability by adding carefully crafted noise into the intermediate layer\nactivations. ANL and CANL can be easily implemented and integrated with most of\nthe mainstream CNN-based models. We compared the effects of the different types\nof noise and visually demonstrate that our proposed adversarial noise instruct\nCNN models to learn to extract cleaner feature maps, which further reduce the\nrisk of over-fitting. We also conclude that models trained with ANL or CANL are\nmore robust to the adversarial examples generated by FGSM than the traditional\nadversarial training approaches.","url_abs":"http://arxiv.org/abs/1805.08000v2","url_pdf":"http://arxiv.org/pdf/1805.08000v2.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-noise-layer-regularize-neural","repo_url":"https://github.com/youzhonghui/ANL","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}