pith. sign in

arxiv: 1808.05933 · v2 · pith:XIYM2DTVnew · submitted 2018-08-17 · 🧮 math.OC · cs.DC· cs.LG

Decentralized Dictionary Learning Over Time-Varying Digraphs

classification 🧮 math.OC cs.DCcs.LG
keywords decentralizedlearningdatadictionaryproblemstime-varyingaggregatingaiming
0
0 comments X
read the original abstract

This paper studies Dictionary Learning problems wherein the learning task is distributed over a multi-agent network, modeled as a time-varying directed graph. This formulation is relevant, for instance, in Big Data scenarios where massive amounts of data are collected/stored in different locations (e.g., sensors, clouds) and aggregating and/or processing all data in a fusion center might be inefficient or unfeasible, due to resource limitations, communication overheads or privacy issues. We develop a unified decentralized algorithmic framework for this class of nonconvex problems, which is proved to converge to stationary solutions at a sublinear rate. The new method hinges on Successive Convex Approximation techniques, coupled with a decentralized tracking mechanism aiming at locally estimating the gradient of the smooth part of the sum-utility. To the best of our knowledge, this is the first provably convergent decentralized algorithm for Dictionary Learning and, more generally, bi-convex problems over (time-varying) (di)graphs.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.