Weiran Lin (Carnegie Mellon University), Keane Lucas (Carnegie Mellon University), Neo Eyal (Tel Aviv University), Lujo Bauer (Carnegie Mellon University), Michael K. Reiter (Duke University), Mahmood Sharif (Tel Aviv University)

Machine-learning models are known to be vulnerable to evasion attacks that perturb model inputs to induce misclassifications. In this work, we identify real-world scenarios where the true threat cannot be assessed accurately by existing attacks. Specifically, we find that conventional metrics measuring targeted and untargeted robustness do not appropriately reflect a model's ability to withstand attacks from one set of source classes to another text set of target classes. To address the shortcomings of existing methods, we formally define a new metric, termed group-based robustness, that complements existing metrics and is better-suited for evaluating model performance in certain attack scenarios. We show empirically that group-based robustness allows us to distinguish between models' vulnerability against specific threat models in situations where traditional robustness metrics do not apply. Moreover, to measure group-based robustness efficiently and accurately, we 1) propose two loss functions and 2) identify three new attack strategies. We show empirically that with comparable success rates, finding evasive samples using our new loss functions saves computation by a factor as large as the number of targeted classes, and finding evasive samples using our new attack strategies saves time by up to 99% compared to brute-force search methods. Finally, we propose a defense method that increases group-based robustness by up to 3.52 times.

View More Papers

GhostType: The Limits of Using Contactless Electromagnetic Interference to...

Qinhong Jiang (Zhejiang University), Yanze Ren (Zhejiang University), Yan Long (University of Michigan), Chen Yan (Zhejiang University), Yumai Sun (University of Michigan), Xiaoyu Ji (Zhejiang University), Kevin Fu (Northeastern University), Wenyuan Xu (Zhejiang University)

Read More

Understanding the Implementation and Security Implications of Protective DNS...

Mingxuan Liu (Zhongguancun Laboratory; Tsinghua University), Yiming Zhang (Tsinghua University), Xiang Li (Tsinghua University), Chaoyi Lu (Tsinghua University), Baojun Liu (Tsinghua University), Haixin Duan (Tsinghua University; Zhongguancun Laboratory), Xiaofeng Zheng (Institute for Network Sciences and Cyberspace, Tsinghua University; QiAnXin Technology Research Institute & Legendsec Information Technology (Beijing) Inc.)

Read More

Decentralized Information-Flow Control for ROS2

Nishit V. Pandya (Indian Institute of Science Bangalore), Himanshu Kumar (Indian Institute of Science Bangalore), Gokulnath M. Pillai (Indian Institute of Science Bangalore), Vinod Ganapathy (Indian Institute of Science Bangalore)

Read More

Merge/Space: A Security Testbed for Satellite Systems

M. Patrick Collins (USC Information Sciences Institute), Alefiya Hussain (USC Information Sciences Institute), J.P. Walters (USC Information Sciences Institute), Calvin Ardi (USC Information Sciences Institute), Chris Tran (USC Information Sciences Institute), Stephen Schwab (USC Information Sciences Institute)

Read More