A forward-backward Bregman splitting scheme for regularized distributed optimization problems

Author(s):  
Jinming Xu ◽  
Shanying Zhu ◽  
Yeng Chai Soh ◽  
Lihua Xie

Author(s):  
Tiep Le ◽  
Tran Cao Son ◽  
Enrico Pontelli

This paper proposes Multi-context System for Optimization Problems (MCS-OP) by introducing conditional costassignment bridge rules to Multi-context Systems (MCS). This novel feature facilitates the definition of a preorder among equilibria, based on the total incurred cost of applied bridge rules. As an application of MCS-OP, the paper describes how MCS-OP can be used in modeling Distributed Constraint Optimization Problems (DCOP), a prominent class of distributed optimization problems that is frequently employed in multi-agent system (MAS) research. The paper shows, by means of an example, that MCS-OP is more expressive than DCOP, and hence, could potentially be useful in modeling distributed optimization problems which cannot be easily dealt with using DCOPs. It also contains a complexity analysis of MCS-OP.





2015 ◽  
Vol 148 ◽  
pp. 278-287 ◽  
Author(s):  
Jianliang Zhang ◽  
Donglian Qi ◽  
Guangzhou Zhao


2018 ◽  
Vol 63 (11) ◽  
pp. 3809-3824 ◽  
Author(s):  
Jinming Xu ◽  
Shanying Zhu ◽  
Yeng Chai Soh ◽  
Lihua Xie


2014 ◽  
Vol 137 (3) ◽  
Author(s):  
Baisravan HomChaudhuri ◽  
Manish Kumar

Distributed optimization methods have been used extensively in multirobot task allocation (MRTA) problems. In distributed optimization domain, most of the algorithms are developed for solving convex optimization problems. However, for complex MRTA problems, the cost function can be nonconvex and multimodal in nature with more than one minimum or maximum points. In this paper, an effort has been made to address these complex MRTA problems with multimodal cost functions in a distributed manner. The approach used in this paper is a distributed primal–dual interior point method where noise is added in the search direction as a mechanism to allow the algorithm to escape from suboptimal solutions. The search direction from the distributed primal–dual interior point method and the weighted variable updates help in the generation of feasible primal and dual solutions and in faster convergence while the noise added in the search direction helps in avoiding local optima. The optimality and the computation time of this proposed method are compared with that of the genetic algorithm (GA) and the numerical results are provided in this paper.



2016 ◽  
Vol 170 (2) ◽  
pp. 493-511 ◽  
Author(s):  
Shreyas Vathul Subramanian ◽  
Daniel A. DeLaurentis ◽  
Dengfeng Sun


Sign in / Sign up

Export Citation Format

Share Document