English

Distributed Gradient Methods with Variable Number of Working Nodes

Information Theory 2016-08-24 v3 math.IT Optimization and Control

Abstract

We consider distributed optimization where NN nodes in a connected network minimize the sum of their local costs subject to a common constraint set. We propose a distributed projected gradient method where each node, at each iteration kk, performs an update (is active) with probability pkp_k, and stays idle (is inactive) with probability 1pk1-p_k. Whenever active, each node performs an update by weight-averaging its solution estimate with the estimates of its active neighbors, taking a negative gradient step with respect to its local cost, and performing a projection onto the constraint set; inactive nodes perform no updates. Assuming that nodes' local costs are strongly convex, with Lipschitz continuous gradients, we show that, as long as activation probability pkp_k grows to one asymptotically, our algorithm converges in the mean square sense (MSS) to the same solution as the standard distributed gradient method, i.e., as if all the nodes were active at all iterations. Moreover, when pkp_k grows to one linearly, with an appropriately set convergence factor, the algorithm has a linear MSS convergence, with practically the same factor as the standard distributed gradient method. Simulations on both synthetic and real world data sets demonstrate that, when compared with the standard distributed gradient method, the proposed algorithm significantly reduces the overall number of per-node communications and per-node gradient evaluations (computational cost) for the same required accuracy.

Keywords

Cite

@article{arxiv.1504.04049,
  title  = {Distributed Gradient Methods with Variable Number of Working Nodes},
  author = {Dusan Jakovetic and Dragana Bajovic and Natasa Krejic and Natasa Krklec-Jerinkic},
  journal= {arXiv preprint arXiv:1504.04049},
  year   = {2016}
}

Comments

submitted to a journal on April 15, 2015; revised on September 23, 2015, and March 10, 2016

R2 v1 2026-06-22T09:16:49.749Z