Dynamic Average Diffusion with randomized Coordinate Updates

Abstract

This work derives and analyzes an online learning strategy for tracking the average of time-varying distributed signals by relying on randomized coordinate-descent updates. During each iteration, each agent selects or observes a random entry of the observation vector, and different agents may select different entries of their observations before engaging in a consultation step. Careful coordination of the interactions among agents is necessary to avoid bias and ensure convergence. We provide a convergence analysis for the proposed methods, and illustrate the results by means of simulations.

References in corpus (14)

• Diffusion Strategies Outperform Consensus Strategies for Distributed Estimation over Adaptive Networks
• ARock: an Algorithmic Framework for Asynchronous Parallel Coordinate Updates
• A decentralized proximal-gradient method with network independent step-sizes and separated convergence rates
• Convergence Rate Analysis of Distributed Gossip (Linear Parameter) Estimation: Fundamental Limits and Tradeoffs
• Distributed Basis Pursuit
• Distributed Pareto Optimization via Diffusion Strategies
• Stochastic Gradient Push for Distributed Deep Learning
• Dictionary Learning over Distributed Models
• Finito: A Faster, Permutable Incremental Gradient Method for Big Data Problems
• Supervised Learning Under Distributed Features
• A Push-Pull Gradient Method for Distributed Optimization in Networks
• Coordinate-Descent Diffusion Learning by Networked Agents
• Exact Diffusion for Distributed Optimization and Learning --- Part I: Algorithm Development
• Accelerated Decentralized Optimization with Local Updates for Smooth and Strongly Convex Objectives