1 citations · 1 across the 1 of their papers we have counts for
2 papers
cs.LG2021★ 1 cited
CD-SGD: Distributed Stochastic Gradient Descent with Compression and Delay Compensation
Enda Yu, Dezun Dong, Yemao Xu +2
Communication overhead is the key challenge for distributed training. Gradient compression is a widely used approach to reduce communication traffic. When combining with parallel c…
cs.DC2020
Communication optimization strategies for distributed deep neural network training: A survey
Shuo Ouyang, Dezun Dong, Yemao Xu +1
Recent trends in high-performance computing and deep learning have led to the proliferation of studies on large-scale deep neural network training. However, the frequent communicat…