paper-with-me

홈 › Papers

DecentLaM: Decentralized Momentum SGD for Large-batch Deep Training

2021-04-24 · ICCV 2021 10 · Kun Yuan, Yiming Chen, Xinmeng Huang, Yingya Zhang, Pan Pan, Yinghui Xu, Wotao Yin

The scale of deep learning nowadays calls for efficient distributed training algorithms. Decentralized momentum SGD (DmSGD), in which each node averages only with its neighbors, is more communication efficient than vanilla Parallel momentum SGD that incurs global average across all computing nodes. On the other hand, the large-batch training has been demonstrated critical to achieve runtime speedup. This motivates us to investigate how DmSGD performs in the large-batch scenario. In this work, we find the momentum term can amplify the inconsistency bias in DmSGD. Such bias becomes more evident as batch-size grows large and hence results in severe performance degradation. We next propose DecentLaM, a novel decentralized large-batch momentum SGD to remove the momentum-incurred bias. The convergence rate for both non-convex and strongly-convex scenarios is established. Our theoretical results justify the superiority of DecentLaM to DmSGD especially in the large-batch scenario. Experimental results on a variety of computer vision tasks and models demonstrate that DecentLaM promises both efficient and high-quality training.

📄 PDF Abstract BibTeX arXiv:2104.11981

Code (1)

Bluefog-Lib/bluefog 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

SGD Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using…

Similar Papers 제목 키워드 기반

MDPGT: Momentum-based Decentralized Policy Gradient Tracking

2021-12-06 · Zhanhong Jiang, Xian Yeow Lee, Sin Yong Tan, Kai Liang Tan 외

We propose a novel policy gradient method for multi-agent reinforcement learning, which leverages two different variance-reduction techniques and does not require large batches over iterations. Specifically, we propose a…

Multi-agent Reinforcement LearningPolicy Gradient Methodsreinforcement-learningReinforcement Learning+2

On Batch Adaptive Training for Deep Learning: Lower Loss and Larger Step Size

2018-01-01 · ICLR 2018 1 · Runyao Chen, Kun Wu, Ping Luo

Mini-batch gradient descent and its variants are commonly used in deep learning. The principle of mini-batch gradient descent is to use noisy gradient calculated on a batch to estimate the real gradient, thus balancing t…

image-classificationImage Classification

Perfect Parallelization in Mini-Batch SGD with Classical Momentum Acceleration

2026-05-18 · Sachin Garg, Michał Dereziński arxiv

Accelerating stochastic gradient methods with classical momentum schemes, such as Polyak's heavy ball, has proven highly successful in training large-scale machine learning models, particularly when combined with the har…

Stochastic Normalized Gradient Descent with Momentum for Large-Batch Training

2020-07-28 · Shen-Yi Zhao, Chang-Wei Shi, Yin-Peng Xie, Wu-Jun Li

Stochastic gradient descent~(SGD) and its variants have been the dominating optimization methods in machine learning. Compared to SGD with small-batch training, SGD with large-batch training can better utilize the comput…

SQuARM-SGD: Communication-Efficient Momentum SGD for Decentralized Optimization

2020-05-13 · Navjot Singh, Deepesh Data, Jemin George, Suhas Diggavi

In this paper, we propose and analyze SQuARM-SGD, a communication-efficient algorithm for decentralized training of large-scale machine learning models over a network. In SQuARM-SGD, each node performs a fixed number of …