# Parallel Restarted SGD with Faster Convergence and Less Communication: Demystifying Why Model Averaging Works for Deep Learning

> Research article (Proceedings of the AAAI Conference on Artificial Intelligence, 2019) · cited 510× · AI/ML

**Wikidata**: [openalex:W2900182564](https://www.wikidata.org/wiki/openalex:W2900182564)  
**Source**: https://4ort.xyz/entity/parallel-restarted-sgd-with-faster-convergence-and-less-communication-demystifying-why-model-averaging-works-for-deep-le