Wanlun Ma (Swinburne University of Technology), Derui Wang (CSIRO’s Data61), Ruoxi Sun (The University of Adelaide & CSIRO's Data61), Minhui Xue (CSIRO's Data61), Sheng Wen (Swinburne University of Technology), Yang Xiang (Digital Research & Innovation Capability Platform, Swinburne University of Technology)

Deep Neural Networks (DNNs) are susceptible to backdoor attacks during training. The model corrupted in this way functions normally, but when triggered by certain patterns in the input, produces a predefined target label. Existing defenses usually rely on the assumption of the universal backdoor setting in which poisoned samples share the same uniform trigger. However,
recent advanced backdoor attacks show that this assumption is no longer valid in dynamic backdoors where the triggers vary from input to input, thereby defeating the existing defenses.

In this work, we propose a novel technique, Beatrix (backdoor detection via Gram matrix). Beatrix utilizes Gram matrix to capture not only the feature correlations but also the appropriately high-order information of the representations. By learning class-conditional statistics from activation patterns of normal samples, Beatrix can identify poisoned samples by capturing the anomalies in activation patterns. To further improve the performance in identifying target labels, Beatrix leverages kernel-based testing without making any prior assumptions on representation distribution. We demonstrate the effectiveness of our method through extensive evaluation and comparison with state-of-the-art defensive techniques. The experimental results show that our approach achieves an F1 score of 91.1% in detecting dynamic backdoors, while the state of the art can only reach 36.9%.

View More Papers

Automatic Retrieval of Privacy Factors from IoMT Policies: ML...

Nyteisha Bookert, Mohd Anwar (North Carolina Agricultural and Technical State University)

Read More

BlockScope: Detecting and Investigating Propagated Vulnerabilities in Forked Blockchain...

Xiao Yi (The Chinese University of Hong Kong), Yuzhou Fang (The Chinese University of Hong Kong), Daoyuan Wu (The Chinese University of Hong Kong), Lingxiao Jiang (Singapore Management University)

Read More

Navigating Murky Waters: Automated Browser Feature Testing for Uncovering...

Mir Masood Ali (University of Illinois Chicago), Binoy Chitale (Stony Brook University), Mohammad Ghasemisharif (University of Illinois Chicago), Chris Kanich (University of Illinois Chicago), Nick Nikiforakis (Stony Brook University), Jason Polakis (University of Illinois Chicago)

Read More

OBSan: An Out-Of-Bound Sanitizer to Harden DNN Executables

Yanzuo Chen (The Hong Kong University of Science and Technology), Yuanyuan Yuan (The Hong Kong University of Science and Technology), Shuai Wang (The Hong Kong University of Science and Technology)

Read More