0038.pdf (305.89 kB)
Download file

Learning optimal temperature region for solving mixed integer functional DCOPs

Download (305.89 kB)
conference contribution
posted on 05.11.2021, 09:59 by Saaduddin Mahmud, Md. Mosaddek Khan, Moumita Choudhury, Long Tran-Thanh, Nick JenningsNick Jennings
Distributed Constraint Optimization Problems (DCOPs) are an important framework for modeling coordinated decision-making problems in multiagent systems with a set of discrete variables. Later works have extended DCOPs to model problems with a set of continuous variables, named Functional DCOPs (F-DCOPs). In this paper, we combine both of these frameworks into the Mixed Integer Functional DCOP (MIF-DCOP) framework that can deal with problems regardless of their variables' type. We then propose a novel algorithm - Distributed Parallel Simulated Annealing (DPSA), where agents cooperatively learn the optimal parameter configuration for the algorithm while also solving the given problem using the learned knowledge. Finally, we empirically evaluate our approach in DCOP, F-DCOP, and MIF-DCOP settings and show that DPSA produces solutions of significantly better quality than the state-of-the-art non-exact algorithms in their corresponding settings.

Funding

ICT Innovation Fund of Bangladesh Government

University Grants Commission (UGC) of Bangladesh

History

Published in

Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence

Pages

268 - 275

Source

Twenty-Ninth International Joint Conference on Artificial Intelligence (IJCAI-20)

Publisher

International Joint Conferences on Artificial Intelligence

Version

VoR (Version of Record)

Rights holder

© International Joint Conferences on Artificial Intelligence

Copyright date

2020

ISBN

9780999241165

ISSN

1045-0823

Language

en

Editor(s)

Christian Bessiere

Location

Yokohama, Japan (Virtual)

Event dates

7th January 2021 - 15th January 2021