Decentralized Proximal Gradient Algorithms With Linear Convergence Rates
From MaRDI portal
Abstract: This work studies a class of non-smooth decentralized multi-agent optimization problems where the agents aim at minimizing a sum of local strongly-convex smooth components plus a common non-smooth term. We propose a general primal-dual algorithmic framework that unifies many existing state-of-the-art algorithms. We establish linear convergence of the proposed method to the exact solution in the presence of the non-smooth term. Moreover, for the more general class of problems with agent specific non-smooth terms, we show that linear convergence cannot be achieved (in the worst case) for the class of algorithms that uses the gradients and the proximal mappings of the smooth and non-smooth parts, respectively. We further provide a numerical counterexample that shows how some state-of-the-art algorithms fail to converge linearly for strongly-convex objectives and different local non-smooth terms.
Cited in
(19)- Distributed decision-coupled constrained optimization via proximal-tracking
- Distributed resource allocation via multi-agent systems under time-varying networks
- Dualize, split, randomize: toward fast nonsmooth optimization algorithms
- Convergence results of a nested decentralized gradient method for non-strongly convex problems
- Distributed composite optimization for multi-agent systems with asynchrony
- Proximal nested primal-dual gradient algorithms for distributed constraint-coupled composite optimization
- On Nonconvex Decentralized Gradient Descent
- scientific article; zbMATH DE number 7626753 (Why is no real title available?)
- Decentralized Accelerated Gradient Methods With Increasing Penalty Parameters
- Graph Topology Invariant Gradient and Sampling Complexity for Decentralized and Stochastic Optimization
- Linear convergence rate analysis of a class of exact first-order distributed methods for weight-balanced time-varying networks and uncoordinated step sizes
- A Unified Framework for Continuous-Time Unconstrained Distributed Optimization
- Recent theoretical advances in decentralized distributed convex optimization
- Online composite optimization with time-varying regularizers
- Balancing communication and computation in gradient tracking algorithms for decentralized optimization
- On graphs with finite-time consensus and their use in gradient tracking
- Local adapt-then-combine algorithms for distributed nonsmooth optimization: achieving provable communication acceleration
- A first-order algorithm for decentralised min-max problems
- Analysis of DGD for decentralized high-dimensional statistical models: beyond the least squares
This page was built for publication: Decentralized Proximal Gradient Algorithms With Linear Convergence Rates
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5002086)
