# Adaptive Gradient Quantization for Data-Parallel SGD

Source: [Adaptive Gradient Quantization for Data-Parallel SGD](<https://www.sota2.com/research/paper/68fa68a11e209131cea1d43c>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-28T13:56:45.132Z

Authors: Fartash Faghri, Iman Tabrizian, Ilia Markov, Dan Alistarh, Daniel Roy, Ali Ramezani-Kebrya

Published: 2020-10-23T14:58:02.000Z

## Abstract

Many communication-efficient variants of SGD use gradient quantization schemes. These schemes are often heuristic and fixed over the course of training. We empirically observe that the statistics of gradients of deep models change during the training. Motivated by this observation, we introduce two adaptive quantization schemes, ALQ and AMQ. In both schemes, processors update their compression schemes in parallel by efficiently computing sufficient statistics of a parametric distribution. We improve the validation accuracy by almost 2% on CIFAR-10 and 1% on ImageNet in challenging low-cost communication setups. Our adaptive methods are also significantly more robust to the choice of hyperparameters.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2010.12460>)

[PDF](<https://arxiv.org/pdf/2010.12460.pdf>)