no code implementations • 28 Feb 2024 • Haoyu Xie, Changqi Wang, Jian Zhao, Yang Liu, Jun Dan, Chong Fu, Baigui Sun
To address this issue, we propose a robust contrastive-based S4 framework, termed the Probabilistic Representation Contrastive Learning (PRCL) framework to enhance the robustness of the unsupervised training process.
no code implementations • ICCV 2023 • Changqi Wang, Haoyu Xie, Yuhui Yuan, Chong Fu, Xiangyu Yue
To improve the robustness of representations, powerful methods introduce a pixel-wise contrastive learning approach in latent space (i. e., representation space) that aggregates the representations to their prototypes in a fully supervised manner.
no code implementations • CVPR 2023 • Xu Zheng, Jinjing Zhu, Yexin Liu, Zidong Cao, Chong Fu, Lin Wang
Moreover, adversarial intra-projection training is proposed to reduce the inherent gap, between the features of the pinhole images and those of the ERP and TP images, respectively.
1 code implementation • 28 Feb 2023 • Chong Fu, Xuhong Zhang, Shouling Ji, Ting Wang, Peng Lin, Yanghe Feng, Jianwei Yin
Thus, in this paper, we propose FreeEagle, the first data-free backdoor detection method that can effectively detect complex backdoor attacks on deep neural networks, without relying on the access to any clean samples or samples with the trigger.
no code implementations • 1 Dec 2022 • Pengyu Qiu, Xuhong Zhang, Shouling Ji, Chong Fu, Xing Yang, Ting Wang
Our work shows that hashing is a promising solution to counter data reconstruction attacks.
1 code implementation • 26 Oct 2022 • Haoyu Xie, Changqi Wang, Mingkai Zheng, Minjing Dong, Shan You, Chong Fu, Chang Xu
In prevalent pixel-wise contrastive learning solutions, the model maps pixels to deterministic representations and regularizes them in the latent space.
2 code implementations • USENIX Security 22 2022 • Chong Fu, Xuhong Zhang, Shouling Ji, Jinyin Chen, Jingzheng Wu, Shanqing Guo, Jun Zhou, Alex X. Liu, Ting Wang
However, we discover that the bottom model structure and the gradient update mechanism of VFL can be exploited by a malicious participant to gain the power to infer the privately owned labels.
no code implementations • 6 Sep 2022 • Xu Zheng, Yunhao Luo, Chong Fu, Kangcheng Liu, Lin Wang
To this end, we propose class-aware feature consistency distillation (CFCD) that first leverages the outputs of each student as the pseudo labels and generates class-aware feature (CF) maps for knowledge transfer between the two students.
1 code implementation • 7 Apr 2022 • Yuhao Mao, Chong Fu, Saizhuo Wang, Shouling Ji, Xuhong Zhang, Zhenguang Liu, Jun Zhou, Alex X. Liu, Raheem Beyah, Ting Wang
To bridge this critical gap, we conduct the first large-scale systematic empirical study of transfer attacks against major cloud-based MLaaS platforms, taking the components of a real transfer attack into account.
no code implementations • 23 Nov 2021 • Xu Zheng, Chong Fu, Haoyu Xie, Jialei Chen, Xingwei Wang, Chiu-Wing Sham
However, due to the scarcity of labeled data, the features extracted by the models are limited in supervised learning, and the quality of predictions for unlabeled data also cannot be guaranteed.
no code implementations • 28 Jun 2021 • Wenchao Zhang, Chong Fu, Xiangshi Chang, Tengfei Zhao, Xiang Li, Chiu-Wing Sham
Without losing generality, we can also build a more lighter head network for other multi-stage detectors by assembling our method.
no code implementations • 1 Jan 2021 • Wenchao Zhang, Haoyu Xie, Mai Zhu, Chong Fu
RoIPool/RoIAlign is an indispensable process for the typical two-stage object detection algorithm, it is used to rescale the object proposal cropped from the feature pyramid to generate a fixed size feature map.
no code implementations • 4 Dec 2020 • Wenchao Zhang, Chong Fu, Haoyu Xie, Mai Zhu, Ming Tie, Junxin Chen
The core component of our GCA framework is a context aware mechanism, in which both global feature pyramid and attention strategies are used for feature extraction and feature refinement, respectively.
no code implementations • 2 Aug 2020 • Wenchao Zhang, Chong Fu, Mai Zhu
The attributes of object contours has great significance for instance segmentation task.
1 code implementation • ICLR 2019 • Mai Zhu, Bo Chang, Chong Fu
A convolutional neural network can be constructed using numerical methods for solving dynamical systems, since the forward pass of the network can be regarded as a trajectory of a dynamical system.