JointSQ: Joint Sparsification-Quantization for Distributed Learning
Authors:
- 16 June 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5778-5787
Abstract
Gradient sparsification and quantization offer a promising prospect to alleviate the communication overhead problem in distributed learning. However, direct combination of the two results in suboptimal solutions, due to the fact that sparsification and quantization haven't been learned together. In this paper, we propose Joint Sparsification-Quantization (JointSQ) inspired by the discovery that sparsification can be treated as 0-bit quantization, regardless of architectures. Specifically, we mathematically formu-late JointSQ as a mixed-precision quantization problem, expanding the solution space. It can be solved by the designed MCKP-Greedy algorithm. Theoretical analysis demon-strates the minimal compression noise of JointSQ, and ex-tensive experiments on various network architectures, including CNN, RNN, and Transformer, also validate this point. Under the introduction of computation overhead consistent with or even lower than previous methods, JointSQ achieves a compression ratio of 1000× on different models while maintaining near-lossless accuracy and brings 1.4× to 2.9× speedup over existing methods.Keywords
This publication has 15 references indexed in Scilit:
- AC-SGD: Adaptively Compressed SGD for Communication-Efficient Distributed LearningIEEE Journal on Selected Areas in Communications, 2022
- Communication Efficient SGD via Gradient Sampling with Bayes Prior2021
- Differentiable Joint Pruning and Quantization for Hardware Efficiency2020
- Automatic Neural Network Compression by Sparsity-Quantization Joint Learning: A Constrained Optimization-Based Approach2020
- Monte Carlo Gradient Quantization2020
- Accelerating Distributed Deep Learning By Adaptive Gradient Quantization2020
- Optimization Methods for Large-Scale Machine LearningSIAM Review, 2018
- Deep Residual Learning for Image Recognition2016
- Multidimensional Knapsack Problems2012
- ImageNet: A large-scale hierarchical image database2009