{"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/towards-robust-detection-of-adversarial","title":"Towards Robust Detection of Adversarial Examples","arxiv_id":"1706.00633","date":"2017-06-02","proceeding":"NeurIPS 2018 12","authors":["Tianyu Pang","Chao Du","Yinpeng Dong","Jun Zhu"],"abstract":"Although the recent progress is substantial, deep learning methods can be\nvulnerable to the maliciously generated adversarial examples. In this paper, we\npresent a novel training procedure and a thresholding test strategy, towards\nrobust detection of adversarial examples. In training, we propose to minimize\nthe reverse cross-entropy (RCE), which encourages a deep network to learn\nlatent representations that better distinguish adversarial examples from normal\nones. In testing, we propose to use a thresholding strategy as the detector to\nfilter out adversarial examples for reliable predictions. Our method is simple\nto implement using standard algorithms, with little extra training cost\ncompared to the common cross-entropy minimization. We apply our method to\ndefend various attacking methods on the widely used MNIST and CIFAR-10\ndatasets, and achieve significant improvements on robust predictions under all\nthe threat models in the adversarial setting.","url_abs":"http://arxiv.org/abs/1706.00633v4","url_pdf":"http://arxiv.org/pdf/1706.00633v4.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":"towards-robust-detection-of-adversarial","repo_url":"https://github.com/P2333/Reverse-Cross-Entropy","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.00633","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}