Decentralized Deep Learning with Arbitrary Communication Compression
read the original 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).
This paper has not been read by Pith yet.
Forward citations
Cited by 2 Pith papers
-
Decentralized Stochastic Subgradient-type Methods with Communication Compression for Nonsmooth Nonconvex Optimization
A unified framework for decentralized stochastic subgradient methods with compressed communication is proposed, proving global convergence for nonsmooth nonconvex objectives via differential inclusions and developing ...
-
Quantized Stochastic Primal-Dual Methods for Distributed Optimization under Relaxed Global Geometry
q-PDGD achieves linear convergence to a neighborhood under RSI with constant step-size and O(1/k) with diminishing steps in stochastic quantized distributed optimization, matching centralized rates.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.