no code implementations • 3 Feb 2024 • Ao Sun, Yuanyuan Yuan, Pingchuan Ma, Shuai Wang
This paper alleviates the information leakage issue by introducing label supervision in concept predication and constructing a hierarchical concept set.
1 code implementation • 11 Oct 2023 • Ziqi Zhang, Chen Gong, Yifeng Cai, Yuanyuan Yuan, Bingyan Liu, Ding Li, Yao Guo, Xiangqun Chen
These solutions, referred to as TEE-Shielded DNN Partition (TSDP), partition a DNN model into two parts, offloading the privacy-insensitive part to the GPU while shielding the privacy-sensitive part within the TEE.
no code implementations • 12 Sep 2023 • Yanzuo Chen, Zhibo Liu, Yuanyuan Yuan, Sihang Hu, Tianxiang Li, Shuai Wang
Defenses have also been proposed to guard model weights.
1 code implementation • 11 Jun 2023 • Yuanyuan Yuan, Shuai Wang, Zhendong Su
We identify two key properties, independence and continuity, that convert the latent space into a precise and analysis-friendly input space representation for certification.
1 code implementation • NeurIPS 2023 • Ao Sun, Pingchuan Ma, Yuanyuan Yuan, Shuai Wang
For computer vision tasks, mainstream pixel-based XAI methods explain DNN decisions by identifying important pixels, and emerging concept-based XAI explore forming explanations with concepts (e. g., a head in an image).
no code implementations • 3 Oct 2022 • Zhibo Liu, Yuanyuan Yuan, Shuai Wang, Xiaofei Xie, Lei Ma
BTD takes DNN executables and outputs full model specifications, including types of DNN operators, network topology, dimensions, and parameters that are (nearly) identical to those of the input models.
no code implementations • 8 Jan 2022 • Qi Pang, Yuanyuan Yuan, Shuai Wang, Wenting Zheng
Vertical federated learning (VFL) system has recently become prominent as a concept to process data distributed across many individual sources without the need to centralize it.
1 code implementation • 9 Dec 2021 • Yuanyuan Yuan, Qi Pang, Shuai Wang
Recent advances in representation learning and perceptual learning inspired us to consider the reconstruction of media inputs from side channel traces as a cross-modality manifold learning task that can be addressed in a unified manner with an autoencoder framework trained to learn the mapping between media inputs and side channel observations.
no code implementations • 6 Dec 2021 • Qi Pang, Yuanyuan Yuan, Shuai Wang
During fuzzing, MDPFuzz decides which mutated state to retain by measuring if it can reduce cumulative rewards or form a new state sequence.
no code implementations • 3 Dec 2021 • Yuanyuan Yuan, Qi Pang, Shuai Wang
In contrast, we discuss the feasibility of mutating real-world media data with provably high DIV and VAL based on manifold.
1 code implementation • 3 Dec 2021 • Yuanyuan Yuan, Qi Pang, Shuai Wang
We demonstrate that NLC is significantly correlated with the diversity of a test suite across a number of tasks (classification and generation) and data formats (image and text).
1 code implementation • CVPR 2021 • Yuanyuan Yuan, Shuai Wang, Mingyue Jiang, Tsong Yueh Chen
MetaVQA checks whether the answer to (i, q) satisfies metamorphic relationships (MRs), denoting perception consistency, with the composed answers of transformed questions and images.
2 code implementations • ICLR 2021 • Yuanyuan Yuan, Shuai Wang, Junping Zhang
Given the ever-growing adoption of machine learning as a service (MLaaS), image analysis software on cloud platforms has been exploited by reconstructing private user images from system side channels.