MLLGSep 18, 2016

On Randomized Distributed Coordinate Descent with Quantized Updates

arXiv:1609.05539v24 citations
Originality Incremental advance
AI Analysis

This work addresses a practical limitation in distributed optimization for scenarios with communication constraints, but it is incremental as it extends existing methods to quantized updates.

The paper tackles the problem of randomized distributed coordinate descent over finite-capacity channels by deriving sufficient conditions on quantization error for convergence, and verifies these conditions experimentally on a linear regression problem.

In this paper, we study the randomized distributed coordinate descent algorithm with quantized updates. In the literature, the iteration complexity of the randomized distributed coordinate descent algorithm has been characterized under the assumption that machines can exchange updates with an infinite precision. We consider a practical scenario in which the messages exchange occurs over channels with finite capacity, and hence the updates have to be quantized. We derive sufficient conditions on the quantization error such that the algorithm with quantized update still converge. We further verify our theoretical results by running an experiment, where we apply the algorithm with quantized updates to solve a linear regression problem.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes