Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/1905.12948 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916189721067520 |
|---|---|
| author | Shi, Chang-Wei Zhao, Shen-Yi Xie, Yin-Peng Gao, Hao Li, Wu-Jun |
| author_facet | Shi, Chang-Wei Zhao, Shen-Yi Xie, Yin-Peng Gao, Hao Li, Wu-Jun |
| contents | With the rapid growth of data, distributed momentum stochastic gradient descent~(DMSGD) has been widely used in distributed learning, especially for training large-scale deep models. Due to the latency and limited bandwidth of the network, communication has become the bottleneck of distributed learning. Communication compression with sparsified gradient, abbreviated as \emph{sparse communication}, has been widely employed to reduce communication cost. All existing works about sparse communication in DMSGD employ local momentum, in which the momentum only accumulates stochastic gradients computed by each worker locally. In this paper, we propose a novel method, called \emph{\underline{g}}lobal \emph{\underline{m}}omentum \emph{\underline{c}}ompression~(GMC), for sparse communication. Different from existing works that utilize local momentum, GMC utilizes global momentum. Furthermore, to enhance the convergence performance when using more aggressive sparsification compressors (e.g., RBGS), we extend GMC to GMC+. We theoretically prove the convergence of GMC and GMC+. To the best of our knowledge, this is the first work that introduces global momentum for sparse communication in distributed learning. Empirical results demonstrate that, compared with the local momentum counterparts, our GMC and GMC+ can achieve higher test accuracy and exhibit faster convergence, especially under non-IID data distribution. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_1905_12948 |
| institution | arXiv |
| publishDate | 2019 |
| record_format | arxiv |
| spellingShingle | Global Momentum Compression for Sparse Communication in Distributed Learning Shi, Chang-Wei Zhao, Shen-Yi Xie, Yin-Peng Gao, Hao Li, Wu-Jun Machine Learning With the rapid growth of data, distributed momentum stochastic gradient descent~(DMSGD) has been widely used in distributed learning, especially for training large-scale deep models. Due to the latency and limited bandwidth of the network, communication has become the bottleneck of distributed learning. Communication compression with sparsified gradient, abbreviated as \emph{sparse communication}, has been widely employed to reduce communication cost. All existing works about sparse communication in DMSGD employ local momentum, in which the momentum only accumulates stochastic gradients computed by each worker locally. In this paper, we propose a novel method, called \emph{\underline{g}}lobal \emph{\underline{m}}omentum \emph{\underline{c}}ompression~(GMC), for sparse communication. Different from existing works that utilize local momentum, GMC utilizes global momentum. Furthermore, to enhance the convergence performance when using more aggressive sparsification compressors (e.g., RBGS), we extend GMC to GMC+. We theoretically prove the convergence of GMC and GMC+. To the best of our knowledge, this is the first work that introduces global momentum for sparse communication in distributed learning. Empirical results demonstrate that, compared with the local momentum counterparts, our GMC and GMC+ can achieve higher test accuracy and exhibit faster convergence, especially under non-IID data distribution. |
| title | Global Momentum Compression for Sparse Communication in Distributed Learning |
| topic | Machine Learning |
| url | https://arxiv.org/abs/1905.12948 |