Abstract
Graph mining tasks arise from many different application domains, rangingfrom social networks, transportation to E-commerce, etc., which have beenreceiving great attention from the theoretical and algorithmic designcommunities in recent years, and there has been some pioneering work employingthe research-rich Reinforcement Learning (RL) techniques to address graph datamining tasks. However, these graph mining methods and RL models are dispersedin different research areas, which makes it hard to compare them. In thissurvey, we provide a comprehensive overview of RL and graph mining methods andgeneralize these methods to Graph Reinforcement Learning (GRL) as a unifiedformulation. We further discuss the applications of GRL methods across variousdomains and summarize the method descriptions, open-source codes, and benchmarkdatasets of GRL methods. Furthermore, we propose important directions andchallenges to be solved in the future. As far as we know, this is the latestwork on a comprehensive survey of GRL, this work provides a global view and alearning resource for scholars. In addition, we create an online open-sourcefor both interested scholars who want to enter this rapidly developing domainand experts who would like to compare GRL methods.