A Compressed Gradient Tracking Method for Decentralized Optimization With Linear Convergence
From MaRDI portal
Abstract: Communication compression techniques are of growing interests for solving the decentralized optimization problem under limited communication, where the global objective is to minimize the average of local cost functions over a multi-agent network using only local computation and peer-to-peer communication. In this paper, we propose a novel compressed gradient tracking algorithm (C-GT) that combines gradient tracking technique with communication compression. In particular, C-GT is compatible with a general class of compression operators that unifies both unbiased and biased compressors. We show that C-GT inherits the advantages of gradient tracking-based algorithms and achieves linear convergence rate for strongly convex and smooth objective functions. Numerical examples complement the theoretical findings and demonstrate the efficiency and flexibility of the proposed algorithm.
Cited in
(6)- Decentralized online strongly convex optimization with general compressors and random disturbances
- A compressed gradient tracking method for distributed aggregative optimization
- Differentially private distributed Nash equilibrium seeking based on gradient tracking and communication compression
- Numerical methods for distributed stochastic compositional optimization problems with aggregative structure
- Differentially private and communication-efficient distributed nonconvex optimization algorithms
- Compressed gradient tracking algorithms for distributed nonconvex optimization
This page was built for publication: A Compressed Gradient Tracking Method for Decentralized Optimization With Linear Convergence
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6077130)
