# Parle: parallelizing stochastic gradient descent

arXiv: https://arxiv.org/abs/1707.00424 · HTML: https://mlsanity.com/paper/1707.00424/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Jul 2017.

## Built on by (most cited first, 2 of 2)

- [The Gradient Convergence Bound of Federated Multi-Agent Reinforcement Learning with Efficient Communication](https://mlsanity.com/paper/2103.13026/) (2021)
- [LSAM: Asynchronous Distributed Training with Landscape-Smoothed Sharpness-Aware Minimization](https://mlsanity.com/paper/2509.03110/) (2025)

## Abstract (arXiv)

We propose a new algorithm called Parle for parallel training of deep networks that converges 2-4x faster than a data-parallel implementation of SGD, while achieving significantly improved error rates that are nearly state-of-the-art on several benchmarks including CIFAR-10 and CIFAR-100, without introducing any additional hyper-parameters. We exploit the phenomenon of flat minima that has been shown to lead to improved generalization error for deep networks. Parle requires very infrequent communication with the parameter server and instead performs more computation on each client, which makes it well-suited to both single-machine, multi-GPU settings and distributed implementations.