# Accordion: Adaptive Gradient Communication via Critical Learning Regime Identification

arXiv: https://arxiv.org/abs/2010.16248 · HTML: https://mlsanity.com/paper/2010.16248/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Oct 2020.

## Built on by (most cited first, 19 of 19)

- [CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Serving](https://mlsanity.com/paper/2310.07240/) (2023)
- [FedPara: Low-Rank Hadamard Product for Communication-Efficient Federated Learning](https://mlsanity.com/paper/2108.06098/) (2021)
- [Pufferfish: Communication-efficient Models At No Extra Cost](https://mlsanity.com/paper/2103.03936/) (2021)
- [AutoFreeze: Automatically Freezing Model Blocks to Accelerate Fine-tuning](https://mlsanity.com/paper/2102.01386/) (2021)
- [Rethinking gradient sparsification as total error minimization](https://mlsanity.com/paper/2108.00951/) (2021)
- [Egeria: Efficient DNN Training with Knowledge-Guided Layer Freezing](https://mlsanity.com/paper/2201.06227/) (2022)
- [Adaptive Compression for Communication-Efficient Distributed Training](https://mlsanity.com/paper/2211.00188/) (2022)
- [GraVAC: Adaptive Compression for Communication-Efficient Distributed DL Training](https://mlsanity.com/paper/2305.12201/) (2023)
- [Critical Learning Periods in Federated Learning](https://mlsanity.com/paper/2109.05613/) (2021)
- [ScaDLES: Scalable Deep Learning over Streaming data at the Edge](https://mlsanity.com/paper/2301.08897/) (2023)
- [Accelerating Distributed ML Training via Selective Synchronization](https://mlsanity.com/paper/2307.07950/) (2023)
- [Flexible Communication for Optimal Distributed Learning over Unpredictable Networks](https://mlsanity.com/paper/2312.02493/) (2023)
- [Kimad: Adaptive Gradient Compression with Bandwidth Awareness](https://mlsanity.com/paper/2312.08053/) (2023)
- [EcoLearn: Optimizing the Carbon Footprint of Federated Learning](https://mlsanity.com/paper/2310.17972/) (2023)
- [Solon: Communication-efficient Byzantine-resilient Distributed Training via Redundant Gradients](https://mlsanity.com/paper/2110.01595/) (2021)
- [DEFT: Exploiting Gradient Norm Difference between Model Layers for Scalable Gradient Sparsification](https://mlsanity.com/paper/2307.03500/) (2023)
- [On Using Large-Batches in Federated Learning](https://mlsanity.com/paper/2509.10537/) (2025)
- [DBLP: Phase-Aware Bounded-Loss Transport for Burst-Resilient Distributed ML Training](https://mlsanity.com/paper/2605.01989/) (2026)
- [CIERA: Cross-Iteration Exponent Reuse for Lossless Allgather in Sharded MoE Training](https://mlsanity.com/paper/2609.04609/) (2026)

## Abstract (arXiv)

Distributed model training suffers from communication bottlenecks due to frequent model updates transmitted across compute nodes. To alleviate these bottlenecks, practitioners use gradient compression techniques like sparsification, quantization, or low-rank updates. The techniques usually require choosing a static compression ratio, often requiring users to balance the trade-off between model accuracy and per-iteration speedup. In this work, we show that such performance degradation due to choosing a high compression ratio is not fundamental. An adaptive compression strategy can reduce communication while maintaining final test accuracy. Inspired by recent findings on critical learning regimes, in which small gradient errors can have irrecoverable impact on model performance, we propose Accordion a simple yet effective adaptive compression algorithm. While Accordion maintains a high enough compression rate on average, it avoids over-compressing gradients whenever in critical learning regimes, detected by a simple gradient-norm based criterion. Our extensive experimental study over a number of machine learning tasks in distributed environments indicates that Accordion, maintains similar model accuracy to uncompressed training, yet achieves up to 5.5x better compression and up to 4.1x end-to-end speedup over static approaches. We show that Accordion also works for adjusting the batch size, another popular strategy for alleviating communication bottlenecks.