Proceedings ArticleDOI
Distributed primal-dual method for multi-agent sharing problem with conic constraints
Necdet Serhat Aybat,Erfan Yazdandoost Hamedani +1 more
- pp 777-782
TLDR
This work considers cooperative multi-agent resource sharing problems over an undirected network of agents, where only those agents connected by an edge can directly communicate, and provides convergence rates in sub-optimality, infeasibility and consensus violation for agents' dual price assessments.Abstract:
We consider cooperative multi-agent resource sharing problems over an undirected network of agents, where only those agents connected by an edge can directly communicate. The objective is to minimize the sum of agent-specific composite convex functions subject to a conic constraint that couples agents' decisions. A distributed primal-dual algorithm is proposed to solve the saddle point formulation, which requires to compute a consensus dual price for the coupling constraint. We provide convergence rates in sub-optimality, infeasibility and consensus violation for agents' dual price assessments; examine the effect of underlying network topology on the convergence rates of the proposed decentralized algorithm; and compare our method with Prox-JADMM algorithm on the basis pursuit problem.read more
Citations
More filters
Journal ArticleDOI
A Distributed ADMM-like Method for Resource Sharing over Time-Varying Networks
TL;DR: It is shown that primal-dual iterate sequence converges to a point defined by a primal optimal solution and a consensual dual price for the coupling constraint and is compared with a centralized method on the basis of denoising and multi-channel power allocation problems.
Posted Content
Improved Convergence Rates for Distributed Resource Allocation
TL;DR: This paper introduces an algorithm for solving the resource allocation problem with an o(1/k) convergence rate when the agents' objective functions are generally convex and per agent local constraints are allowed and introduces a gradient-based algorithm which achieves geometric convergence with an improved scalability.
Proceedings ArticleDOI
Decentralized Resource Allocation via Dual Consensus ADMM
TL;DR: In this paper, the authors consider a resource allocation problem over an undirected network of agents, where edges of the network define communication links, and derive two methods by applying the alternating direction method of multipliers (ADMM) for decentralized consensus optimization.
Journal ArticleDOI
Distributed Resource Allocation Over Dynamic Networks With Uncertainty
Thinh T. Doan,Carolyn L. Beck +1 more
TL;DR: In this article, a distributed subgradient method is proposed to solve the dual problem of resource allocation in large-scale networks with complex interconnection structures, where any solution must be implemented in parallel and based only on local data resulting in a need for distributed algorithms.
Posted Content
A distributed ADMM-like method for resource sharing under conic constraints over time-varying networks
TL;DR: A distributed primal-dual algorithm DPDA-D is proposed to solve the saddle point formulation of the sharing problem on time-varying (un)directed communication networks; and it is shown that primal- dual iterate sequence converges to a point defined by a primal optimal solution and a consensual dual price for the coupling constraint.
References
More filters
Journal ArticleDOI
A First-Order Primal-Dual Algorithm for Convex Problems with Applications to Imaging
Antonin Chambolle,Thomas Pock +1 more
TL;DR: A first-order primal-dual algorithm for non-smooth convex optimization problems with known saddle-point structure can achieve O(1/N2) convergence on problems, where the primal or the dual objective is uniformly convex, and it can show linear convergence, i.e. O(ωN) for some ω∈(0,1), on smooth problems.
Journal ArticleDOI
Subgradient Methods for Saddle-Point Problems
Angelia Nedic,Asuman Ozdaglar +1 more
TL;DR: This work presents a subgradient algorithm for generating approximate saddle points and provides per-iteration convergence rate estimates on the constructed solutions, and focuses on Lagrangian duality, where it is shown this algorithm is particularly well-suited for problems where the subgradient of the dual function cannot be evaluated easily.
Journal ArticleDOI
On the ergodic convergence rates of a first-order primal---dual algorithm
Antonin Chambolle,Thomas Pock +1 more
TL;DR: The proofs of convergence for a first order primal–dual algorithm for convex optimization is revisited, with simpler proofs and more complete results that can deal with explicit terms and nonlinear proximity operators in spaces with quite general norms.
Journal ArticleDOI
Parallel Multi-Block ADMM with o(1 / k) Convergence
TL;DR: The classic ADMM can be extended to the N-block Jacobi fashion and preserve convergence in the following two cases: (i) matrices A_i and Ai are mutually near-orthogonal and have full column-rank, or (ii) proximal terms are added to theN subproblems (but without any assumption on matrices $$A_i$$Ai).
Journal ArticleDOI
Multi-Agent Distributed Optimization via Inexact Consensus ADMM
TL;DR: Low-complexity algorithms are proposed that can reduce the overall computational cost of consensus ADMM by an order of magnitude for certain large-scale problems and offer considerably lower computational complexity.