# AdaGossip: Adaptive Consensus Step-size for Decentralized Deep Learning with Communication Compression

arXiv: https://arxiv.org/abs/2404.05919 · HTML: https://mlsanity.com/paper/2404.05919/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 26 references that are in the map.
- First posted on arXiv 9 Apr 2024.

## Builds on (most cited first, 20 of 20)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms](https://mlsanity.com/paper/1708.07747/) (2017)
- [Federated Optimization: Distributed Machine Learning for On-Device Intelligence](https://mlsanity.com/paper/1610.02527/) (2016)
- [Can Decentralized Algorithms Outperform Centralized Algorithms? A Case Study for Decentralized Parallel Stochastic Gradient Descent](https://mlsanity.com/paper/1705.09056/) (2017)
- [PowerSGD: Practical Low-Rank Gradient Compression for Distributed Optimization](https://mlsanity.com/paper/1905.13727/) (2019)
- [Decentralized Stochastic Optimization and Gossip Algorithms with Compressed Communication](https://mlsanity.com/paper/1902.00340/) (2019)
- [Stochastic Gradient Push for Distributed Deep Learning](https://mlsanity.com/paper/1811.10792/) (2018)
- [Decentralized Deep Learning with Arbitrary Communication Compression](https://mlsanity.com/paper/1907.09356/) (2019)
- [Quasi-Global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data](https://mlsanity.com/paper/2102.04761/) (2021)
- [Evolving Normalization-Activation Layers](https://mlsanity.com/paper/2004.02967/) (2020)
- [BEER: Fast O(1/T) Rate for Decentralized Nonconvex Optimization with Communication Compression](https://mlsanity.com/paper/2201.13320/) (2022)
- [PowerGossip: Practical Low-Rank Communication Compression in Decentralized Deep Learning](https://mlsanity.com/paper/2008.01425/) (2020)
- [Quantized Decentralized Stochastic Learning over Directed Graphs](https://mlsanity.com/paper/2002.09964/) (2020)
- [Moniqua: Modulo Quantized Communication in Decentralized SGD](https://mlsanity.com/paper/2002.11787/) (2020)
- [textttDeepSqueeze: Decentralization Meets Error-Compensated Compression](https://mlsanity.com/paper/1907.07346/) (2019)
- [Sparse-Push: Communication- & Energy-Efficient Decentralized Distributed Learning over Directed & Time-Varying Graphs with non-IID Datasets](https://mlsanity.com/paper/2102.05715/) (2021)
- [Low Precision Decentralized Distributed Training over IID and non-IID Data](https://mlsanity.com/paper/2111.09389/) (2021)

## Built on by (most cited first, 1 of 1)

- [SeedFlood: A Step Toward Scalable Decentralized Fine-Tuning of LLMs](https://mlsanity.com/paper/2602.18181/) (2026)

## Abstract (arXiv)

Decentralized learning is crucial in supporting on-device learning over large distributed datasets, eliminating the need for a central server. However, the communication overhead remains a major bottleneck for the practical realization of such decentralized setups. To tackle this issue, several algorithms for decentralized training with compressed communication have been proposed in the literature. Most of these algorithms introduce an additional hyper-parameter referred to as consensus step-size which is tuned based on the compression ratio at the beginning of the training. In this work, we propose AdaGossip, a novel technique that adaptively adjusts the consensus step-size based on the compressed model differences between neighboring agents. We demonstrate the effectiveness of the proposed method through an exhaustive set of experiments on various Computer Vision datasets (CIFAR-10, CIFAR-100, Fashion MNIST, Imagenette, and ImageNet), model architectures, and network topologies. Our experiments show that the proposed method achieves superior performance (0-2% improvement in test accuracy) compared to the current state-of-the-art method for decentralized learning with communication compression.