A Distributed Stochastic Gradient Tracking Method
Authors:
- 1 December 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 963-968
Abstract
In this paper, we study the problem of distributed multi-agent optimization over a network, where each agent possesses a local cost function that is smooth and strongly convex. The global objective is to find a common solution that minimizes the average of all cost functions. Assuming agents only have access to unbiased estimates of the gradients of their local cost functions, we consider a distributed stochastic gradient tracking method. We show that, in expectation, the iterates generated by each agent are attracted to a neighborhood of the optimal solution, where they accumulate exponentially fast (under a constant step size choice). More importantly, the limiting (expected) error bounds on the distance of the iterates from the optimal solution decrease with the network size, which is a comparable performance to a centralized stochastic gradient algorithm. Numerical examples further demonstrate the effectiveness of the method.Keywords
All Related Versions
This publication has 27 references indexed in Scilit:
- Distributed convex optimization via continuous-time coordination algorithms with discrete-time communicationAutomatica, 2015
- Cooperative Distributed Optimization in Multiagent Networks With DelaysIEEE Transactions on Systems, Man, and Cybernetics: Systems, 2014
- Adaptive Penalty-Based Distributed Stochastic Convex OptimizationIEEE Transactions on Signal Processing, 2014
- Proximal-Gradient Algorithms for Tracking Cascades Over Social NetworksIEEE Journal of Selected Topics in Signal Processing, 2014
- Fast Distributed Gradient MethodsIEEE Transactions on Automatic Control, 2014
- Distributed Stochastic Subgradient Projection Algorithms for Convex OptimizationJournal of Optimization Theory and Applications, 2010
- Multi-fidelity optimization via surrogate modellingProceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2007
- Distributed asynchronous deterministic and stochastic gradient optimization algorithmsIEEE Transactions on Automatic Control, 1986
- Stochastic Estimation of the Maximum of a Regression FunctionThe Annals of Mathematical Statistics, 1952
- A Stochastic Approximation MethodThe Annals of Mathematical Statistics, 1951