TY - GEN
T1 - New algorithms for continuous distributed constraint optimization problems
AU - Hoang, Khoi D.
AU - Yeoh, William
AU - Yokoo, Makoto
AU - Rabinovich, Zinovi
N1 - Funding Information:
Hoang and Yeoh are partially supported by NSF grants 1812619 and 1838364, Yokoo is partially supported by JSPS KAKENHI grant JP17H00761, and Rabinovich is supported by Singapore MoE AcRF Tier-1 RG24/18 (S).
PY - 2020
Y1 - 2020
N2 - Distributed Constraint Optimization Problems (DCOPs) are a powerful tool to model multi-agent coordination problems that are distributed by nature. The formulation is suitable for problems where variables are discrete and constraint utilities are represented in tabular form. However, many real-world applications have variables that are continuous and tabular forms thus cannot accurately represent constraint utilities. To overcome this limitation, researchers have proposed the Continuous DCOP (C-DCOP) model, which are DCOPs with continuous variables. But existing approaches usually come with some restrictions on the form of constraint utilities and are without quality guarantees. Therefore, in this paper, we (i) propose an exact algorithm to solve a specific subclass of C-DCOPs; (ii) propose an approximation method with quality guarantees to solve general C-DCOPs; (iii) propose additional C-DCOP algorithms that are more scalable; and (iv) empirically show that our algorithms outperform existing state-of-the-art C-DCOP algorithms when given the same communication limitations.
AB - Distributed Constraint Optimization Problems (DCOPs) are a powerful tool to model multi-agent coordination problems that are distributed by nature. The formulation is suitable for problems where variables are discrete and constraint utilities are represented in tabular form. However, many real-world applications have variables that are continuous and tabular forms thus cannot accurately represent constraint utilities. To overcome this limitation, researchers have proposed the Continuous DCOP (C-DCOP) model, which are DCOPs with continuous variables. But existing approaches usually come with some restrictions on the form of constraint utilities and are without quality guarantees. Therefore, in this paper, we (i) propose an exact algorithm to solve a specific subclass of C-DCOPs; (ii) propose an approximation method with quality guarantees to solve general C-DCOPs; (iii) propose additional C-DCOP algorithms that are more scalable; and (iv) empirically show that our algorithms outperform existing state-of-the-art C-DCOP algorithms when given the same communication limitations.
UR - http://www.scopus.com/inward/record.url?scp=85096636243&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85096636243&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85096636243
T3 - Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS
SP - 502
EP - 510
BT - Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2020
A2 - An, Bo
A2 - El Fallah Seghrouchni, Amal
A2 - Sukthankar, Gita
PB - International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS)
T2 - 19th International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2020
Y2 - 19 May 2020
ER -