{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/the-n-tuple-bandit-evolutionary-algorithm-for-1","title":"The N-Tuple Bandit Evolutionary Algorithm for Automatic Game Improvement","arxiv_id":"1705.01080","date":"2017-03-18","proceeding":null,"authors":["Kamolwan Kunanusont","Raluca D. Gaina","Jialin Liu","Diego Perez-Liebana","Simon M. Lucas"],"abstract":"This paper describes a new evolutionary algorithm that is especially well\nsuited to AI-Assisted Game Design. The approach adopted in this paper is to use\nobservations of AI agents playing the game to estimate the game's quality. Some\nof best agents for this purpose are General Video Game AI agents, since they\ncan be deployed directly on a new game without game-specific tuning; these\nagents tend to be based on stochastic algorithms which give robust but noisy\nresults and tend to be expensive to run. This motivates the main contribution\nof the paper: the development of the novel N-Tuple Bandit Evolutionary\nAlgorithm, where a model is used to estimate the fitness of unsampled points\nand a bandit approach is used to balance exploration and exploitation of the\nsearch space. Initial results on optimising a Space Battle game variant suggest\nthat the algorithm offers far more robust results than the Random Mutation Hill\nClimber and a Biased Mutation variant, which are themselves known to offer\ncompetitive performance across a range of problems. Subjective observations are\nalso given by human players on the nature of the evolved games, which indicate\na preference towards games generated by the N-Tuple algorithm.","url_abs":"http://arxiv.org/abs/1705.01080v1","url_pdf":"http://arxiv.org/pdf/1705.01080v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"the-n-tuple-bandit-evolutionary-algorithm-for-1","repo_url":"https://github.com/Bam4d/NTBEA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"the-n-tuple-bandit-evolutionary-algorithm-for-1","repo_url":"https://github.com/SimonLucas/ntbea","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"game-design","task_name":"Game Design"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}