Graph-Supported Dynamic Algorithm Configuration for Multi-Objective Combinatorial Optimization

TL;DR

This paper introduces a graph neural network-based deep reinforcement learning approach for dynamic algorithm configuration in multi-objective combinatorial optimization, improving adaptability and generalizability over existing methods.

Contribution

It proposes a novel GNN-based DRL framework modeling dynamic configuration as a Markov decision process for MOCO problems.

Findings

Outperforms traditional and DRL-based methods in efficacy and adaptability.

Shows strong generalizability across different objective types and problem sizes.

Applicable to various evolutionary computation methods.

Abstract

Deep reinforcement learning (DRL) has been widely used for dynamic algorithm configuration, particularly in evolutionary computation, which benefits from the adaptive update of parameters during the algorithmic execution. However, applying DRL to algorithm configuration for multi-objective combinatorial optimization (MOCO) problems remains relatively unexplored. This paper presents a novel graph neural network (GNN) based DRL to configure multi-objective evolutionary algorithms. We model the dynamic algorithm configuration as a Markov decision process, representing the convergence of solutions in the objective space by a graph, with their embeddings learned by a GNN to enhance the state representation. Experiments on diverse MOCO challenges indicate that our method outperforms traditional and DRL-based algorithm configuration methods in terms of efficacy and adaptability. It also exhibits advantageous generalizability across objective types and problem sizes, and applicability to different evolutionary computation methods.