# DoubleSqueeze: Parallel Stochastic Gradient Descent with Double-Pass Error-Compensated Compression

arXiv: https://arxiv.org/abs/1905.05957 · HTML: https://mlsanity.com/paper/1905.05957/

## Key facts

- Cited by 99 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 May 2019.

## Built on by (most cited first, 30 of 99)

- [A Field Guide to Federated Optimization](https://mlsanity.com/paper/2107.06917/) (2021)
- [EF21: A New, Simpler, Theoretically Better, and Practically Faster Error Feedback](https://mlsanity.com/paper/2106.05203/) (2021)
- [Pufferfish: Communication-efficient Models At No Extra Cost](https://mlsanity.com/paper/2103.03936/) (2021)
- [Exponential Graph is Provably Efficient for Decentralized Deep Training](https://mlsanity.com/paper/2110.13363/) (2021)
- [DecentLaM: Decentralized Momentum SGD for Large-batch Deep Training](https://mlsanity.com/paper/2104.11981/) (2021)
- [A Guide Through the Zoo of Biased SGD](https://mlsanity.com/paper/2305.16296/) (2023)
- [FedML-HE: An Efficient Homomorphic-Encryption-Based Privacy-Preserving Federated Learning System](https://mlsanity.com/paper/2303.10837/) (2023)
- [Cross-Gradient Aggregation for Decentralized Learning from Non-IID data](https://mlsanity.com/paper/2103.02051/) (2021)
- [Lower Bounds and Nearly Optimal Algorithms in Distributed Learning with Communication Compression](https://mlsanity.com/paper/2206.03665/) (2022)
- [Permutation Compressors for Provably Faster Distributed Nonconvex Optimization](https://mlsanity.com/paper/2110.03300/) (2021)
- [Momentum Provably Improves Error Feedback!](https://mlsanity.com/paper/2305.15155/) (2023)
- [Federated Learning Under Intermittent Client Availability and Time-Varying Communication Constraints](https://mlsanity.com/paper/2205.06730/) (2022)
- [Fine-tuning Language Models over Slow Networks using Activation Compression with Guarantees](https://mlsanity.com/paper/2206.01299/) (2022)
- [DoCoFL: Downlink Compression for Cross-Device Federated Learning](https://mlsanity.com/paper/2302.00543/) (2023)
- [ProgFed: Effective, Communication, and Computation Efficient Federated Learning by Progressive Training](https://mlsanity.com/paper/2110.05323/) (2021)
- [3PC: Three Point Compressors for Communication-Efficient Distributed Training and a Better Theory for Lazy Aggregation](https://mlsanity.com/paper/2202.00998/) (2022)
- [Time-Correlated Sparsification for Communication-Efficient Federated Learning](https://mlsanity.com/paper/2101.08837/) (2021)
- [On the Utility of Gradient Compression in Distributed Training Systems](https://mlsanity.com/paper/2103.00543/) (2021)
- [Distributed Second Order Methods with Fast Rates and Compressed Communication](https://mlsanity.com/paper/2102.07158/) (2021)
- [Accelerating Gossip SGD with Periodic Global Averaging](https://mlsanity.com/paper/2105.09080/) (2021)
- [1-bit LAMB: Communication Efficient Large-Scale Large-Batch Training with LAMB's Convergence Speed](https://mlsanity.com/paper/2104.06069/) (2021)
- [The Internet of Federated Things (IoFT): A Vision for the Future and In-depth Survey of Data-driven Approaches for Federated Learning](https://mlsanity.com/paper/2111.05326/) (2021)
- [Revisiting Optimal Convergence Rate for Smooth and Non-convex Stochastic Decentralized Optimization](https://mlsanity.com/paper/2210.07863/) (2022)
- [ScaleCom: Scalable Sparsified Gradient Compression for Communication-Efficient Distributed Training](https://mlsanity.com/paper/2104.11125/) (2021)
- [Smoothness Matrices Beat Smoothness Constants: Better Communication Compression Techniques for Distributed Optimization](https://mlsanity.com/paper/2102.07245/) (2021)
- [Preserved central model for faster bidirectional compression in distributed settings](https://mlsanity.com/paper/2102.12528/) (2021)
- [Communication Acceleration of Local Gradient Methods via an Accelerated Primal-Dual Algorithm with Inexact Prox](https://mlsanity.com/paper/2207.03957/) (2022)
- [Distributed Learning Systems with First-order Methods](https://mlsanity.com/paper/2104.05245/) (2021)
- [On the Convergence of Decentralized Adaptive Gradient Methods](https://mlsanity.com/paper/2109.03194/) (2021)
- [Persia: An Open, Hybrid System Scaling Deep Learning-based Recommenders up to 100 Trillion Parameters](https://mlsanity.com/paper/2111.05897/) (2021)

## Abstract (arXiv)

A standard approach in large scale machine learning is distributed stochastic gradient training, which requires the computation of aggregated stochastic gradients over multiple nodes on a network. Communication is a major bottleneck in such applications, and in recent years, compressed stochastic gradient methods such as QSGD (quantized SGD) and sparse SGD have been proposed to reduce communication. It was also shown that error compensation can be combined with compression to achieve better convergence in a scheme that each node compresses its local stochastic gradient and broadcast the result to all other nodes over the network in a single pass. However, such a single pass broadcast approach is not realistic in many practical implementations. For example, under the popular parameter server model for distributed learning, the worker nodes need to send the compressed local gradients to the parameter server, which performs the aggregation. The parameter server has to compress the aggregated stochastic gradient again before sending it back to the worker nodes. In this work, we provide a detailed analysis on this two-pass communication model and its asynchronous parallel variant, with error-compensated compression both on the worker nodes and on the parameter server. We show that the error-compensated stochastic gradient algorithm admits three very nice properties: 1) it is compatible with an arbitrary compression technique; 2) it admits an improved convergence rate than the non error-compensated stochastic gradient methods such as QSGD and sparse SGD; 3) it admits linear speedup with respect to the number of workers. The empirical study is also conducted to validate our theoretical results.