Multi-Agent Election-Based Hyper-Heuristics

Multi-Agent Election-Based Hyper-Heuristics

Vinicius Renan de Carvalho, Jaime Simão Sichman

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence
Doctoral Consortium. Pages 5779-5780. https://doi.org/10.24963/ijcai.2018/833

Hyper-heuristics are high-level methodologies responsible for automatically discover how to combine elements from a low-level heuristic set in order to solve optimization problems. Agents, in turn, are autonomous component responsible for watching an environment and perform some actions according to their perceptions. Thus, agent-based techniques seem suitable for the design of hyper-heuristics. This work presents an agent-based hyper-heuristic framework for choosing the best low-level heuristic. The proposed framework performs a cooperative voting procedure, considering a set of quality indicator voters, to define which multi-objective evolutionary algorithm (MOEA) should generate more new solutions along the execution.
Keywords:
Agent-based and Multi-agent Systems: Coordination and cooperation
Combinatorial & Heuristic Search: Meta-Reasoning and Meta-heuristics
Combinatorial & Heuristic Search: Combinatorial search/optimisation
Agent-based and Multi-agent Systems: Social Choice Theory