A Concentration Bound for Distributed Stochastic Approximation
Harsh Dolhare, Vivek Borkar
TL;DR
This paper provides a high probability bound on the tracking error in distributed stochastic approximation with consensus, using the ODE approach to analyze the classical model.
Contribution
It introduces a novel high probability bound analysis for the classical distributed stochastic approximation scheme with consensus, using the ODE method.
Findings
High probability tracking error bound established
Analysis based on the ODE approach to stochastic approximation
Provides insights into the convergence behavior of distributed schemes
Abstract
We revisit the classical model of Tsitsiklis, Bertsekas and Athans for distributed stochastic approximation with consensus. The main result is an analysis of this scheme using the ODE approach to stochastic approximation, leading to a high probability bound for the tracking error between suitably interpolated iterates and the limiting differential equation. Several future directions will also be highlighted.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsStochastic Gradient Optimization Techniques · Stochastic processes and financial applications · Markov Chains and Monte Carlo Methods