Graph Reinforcement Learning for Combinatorial Optimization: A Survey and Unifying Perspective

arXiv:2404.06492v1 Announce Type: new
Abstract: Graphs are a natural representation for systems based on relations between connected entities. Combinatorial optimization problems, which arise when considering an objective function related to a process of interest on discrete structures, are often challenging due to the rapid growth of the solution space. The trial-and-error paradigm of Reinforcement Learning has recently emerged as a promising alternative to traditional methods, such as exact algorithms and (meta)heuristics, for discovering better decision-making strategies in a variety of …

abstract arxiv cs.ai cs.lg function graph graphs growth natural optimization perspective process reinforcement reinforcement learning relations representation solution space survey systems type