Beyond Traditional Threats: A Persistent Backdoor Attack on Federated Learning

1 code implementation • 26 Apr 2024 • Tao Liu, Yuhang Zhang, Zhu Feng, Zhiqin Yang, Chen Xu, Dapeng Man, Wu Yang

Trained backdoored global model is more resilient to benign updates, leading to a higher attack success rate on the test set.

Backdoor Attack Federated Learning

Bridging the Gaps of Both Modality and Language: Synchronous Bilingual CTC for Speech Translation and Speech Recognition

1 code implementation • 21 Sep 2023 • Chen Xu, Xiaoqian Liu, Erfeng He, Yuhao Zhang, Qianqian Dong, Tong Xiao, Jingbo Zhu, Dapeng Man, Wu Yang

In this study, we present synchronous bilingual Connectionist Temporal Classification (CTC), an innovative framework that leverages dual CTC to bridge the gaps of both modality and language in the speech translation (ST) task.

speech-recognition Speech Recognition +1

TRM-UAP: Enhancing the Transferability of Data-Free Universal Adversarial Perturbation via Truncated Ratio Maximization

1 code implementation • ICCV 2023 • Yiran Liu, Xin Feng, Yunlong Wang, Wu Yang, Di Ming

Aiming at crafting a single universal adversarial perturbation (UAP) to fool CNN models for various data samples, universal attack enables a more efficient and accurate evaluation for the robustness of CNN models.

Prespecified-time observer-based distributed control of battery energy storage systems

no code implementations • 29 Jun 2022 • Wu Yang, Shu-Ming Liang, Yan-Wu Wang, Zhi-Wei Liu

This paper studies the state-of-charge (SoC) balancing and the total charging/discharging power tracking issues for battery energy storage systems (BESSs) with multiple distributed heterogeneous battery units.

Dynamic Ensemble Active Learning: A Non-Stationary Bandit with Expert Advice

no code implementations • 29 Sep 2018 • Pang Kunkun, Dong Mingzhi, Wu Yang, Hospedales Timothy M.

In this paper, we propose dynamic ensemble active learning as a more general and promising research direction.

Active Learning