no code implementations • 2 May 2024 • Chris Xing Tian, Yibing Liu, Haoliang Li, Ray C. C. Cheung, Shiqi Wang
However, FL also faces challenges such as high computational and communication costs regarding resource-constrained devices, and poor generalization performance due to the heterogeneity of data across edge clients and the presence of out-of-distribution data.
no code implementations • 26 Jul 2023 • Jingxin Zhang, Jiawei Xi, Peixing Li, Ray C. C. Cheung, Alex M. H. Wong, Jensen Li
Enabled by the tunability of a programmable metasurface, large sets of experimental data in various configurations can be collected for DNN training.
1 code implementation • ICCV 2023 • Wing-Yin Yu, Lai-Man Po, Ray C. C. Cheung, Yuzhi Zhao, Yu Xue, Kun Li
To address these issues, we propose a novel Deformable Motion Modulation (DMM) that utilizes geometric kernel offset with adaptive weight modulation to simultaneously perform feature alignment and style transfer.
1 code implementation • ICLR 2020 • Junjie Liu, Zhe Xu, Runbin Shi, Ray C. C. Cheung, Hayden K. -H. So
We present a novel network pruning algorithm called Dynamic Sparse Training that can jointly find the optimal network parameters and sparse network structure in a unified optimization process with trainable pruning thresholds.
no code implementations • 25 Sep 2019 • Zhe Xu, Ray C. C. Cheung
Recently, binary convolutional neural networks are explored to help alleviate this issue by quantizing both weights and activations with only 1 single bit.
no code implementations • 17 May 2018 • Zhe Xu, Biao Min, Ray C. C. Cheung
Scene background initialization allows the recovery of a clear image without foreground objects from a video sequence, which is generally the first step in many computer vision and video processing applications.