Communication Efficient Distributed Optimization using an Approximate Newton-type Method
Machine Learning
2014-05-15 v4 Optimization and Control
Machine Learning
Abstract
We present a novel Newton-type method for distributed optimization, which is particularly well suited for stochastic optimization and learning problems. For quadratic objectives, the method enjoys a linear rate of convergence which provably \emph{improves} with the data size, requiring an essentially constant number of iterations under reasonable assumptions. We provide theoretical and empirical evidence of the advantages of our method compared to other approaches, such as one-shot parameter averaging and ADMM.
Cite
@article{arxiv.1312.7853,
title = {Communication Efficient Distributed Optimization using an Approximate Newton-type Method},
author = {Ohad Shamir and Nathan Srebro and Tong Zhang},
journal= {arXiv preprint arXiv:1312.7853},
year = {2014}
}