Back to Search Start Over

Decentralized Deep Learning with Arbitrary Communication Compression

Authors :
Koloskova, Anastasiia
Lin, Tao
Stich, Sebastian Urban
Jaggi, Martin
Publication Year :
2019
Publisher :
arXiv, 2019.

Abstract

Decentralized training of deep learning models is a key element for enabling data privacy and on-device learning over networks, as well as for efficient scaling to large compute clusters. As current approaches suffer from limited bandwidth of the network, we propose the use of communication compression in the decentralized training context. We show that Choco-SGD $-$ recently introduced and analyzed for strongly-convex objectives only $-$ converges under arbitrary high compression ratio on general non-convex functions at the rate $O\bigl(1/\sqrt{nT}\bigr)$ where $T$ denotes the number of iterations and $n$ the number of workers. The algorithm achieves linear speedup in the number of workers and supports higher compression than previous state-of-the art methods. We demonstrate the practical performance of the algorithm in two key scenarios: the training of deep learning models (i) over distributed user devices, connected by a social network and (ii) in a datacenter (outperforming all-reduce time-wise).

Details

Database :
OpenAIRE
Accession number :
edsair.doi.dedup.....5c66b4653a93d5efaf52aca69ee02d76
Full Text :
https://doi.org/10.48550/arxiv.1907.09356