no code implementations • 3 Apr 2024 • Rishub Tamirisa, Chulin Xie, Wenxuan Bao, Andy Zhou, Ron Arel, Aviv Shamsian
Recent methods addressed the client data heterogeneity issue via personalized federated learning (PFL) - a class of FL algorithms aiming to personalize learned global knowledge to better suit the clients' local data distributions.
1 code implementation • NeurIPS 2023 • Wenxuan Bao, Tianxin Wei, Haohan Wang, Jingrui He
To tackle this challenge, we propose a novel algorithm called ATP to adaptively learns the adaptation rates for each module in the model from distribution shifts among source domains.
no code implementations • 10 Jul 2023 • Dongqi Fu, Wenxuan Bao, Ross Maciejewski, Hanghang Tong, Jingrui He
We systematically review related works from the data to the computational aspects.
1 code implementation • 10 Jun 2023 • Wenxuan Bao, Haohan Wang, Jun Wu, Jingrui He
In federated learning (FL), multiple clients collaborate to train machine learning models together while keeping their data decentralized.
no code implementations • 21 May 2023 • Ziqi Wang, Chi Han, Wenxuan Bao, Heng Ji
However, such data augmentation methods are sub-optimal for knowledge distillation since the teacher model could provide label distributions and is more tolerant to semantic shifts.
1 code implementation • 27 Aug 2022 • Wenxuan Bao, Jun Wu, Jingrui He
In federated learning, most existing robust aggregation rules (AGRs) combat Byzantine attacks in the IID setting, where client data is assumed to be independent and identically distributed.
no code implementations • 13 May 2022 • Wenxuan Bao, Luke A. Bauer, Vincent Bindschaedler
The use of differentially private learning algorithms in a "drop-in" fashion -- without accounting for the impact of differential privacy (DP) noise when choosing what feature engineering operations to use, what features to select, or what neural network architecture to use -- yields overly complex and poorly performing models.
no code implementations • CVPR 2020 • Zhiheng Li, Wenxuan Bao, Jiayang Zheng, Chenliang Xu
The perceptual-based grouping process produces a hierarchical and compositional image representation that helps both human and machine vision systems recognize heterogeneous visual concepts.