{"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/tstarbots-defeating-the-cheating-level","title":"TStarBots: Defeating the Cheating Level Builtin AI in StarCraft II in the Full Game","arxiv_id":"1809.07193","date":"2018-09-19","proceeding":null,"authors":["Peng Sun","Xinghai Sun","Lei Han","Jiechao Xiong","Qing Wang","Bo Li","Yang Zheng","Ji Liu","Yongsheng Liu","Han Liu","Tong Zhang"],"abstract":"Starcraft II (SC2) is widely considered as the most challenging Real Time\nStrategy (RTS) game. The underlying challenges include a large observation\nspace, a huge (continuous and infinite) action space, partial observations,\nsimultaneous move for all players, and long horizon delayed rewards for local\ndecisions. To push the frontier of AI research, Deepmind and Blizzard jointly\ndeveloped the StarCraft II Learning Environment (SC2LE) as a testbench of\ncomplex decision making systems. SC2LE provides a few mini games such as\nMoveToBeacon, CollectMineralShards, and DefeatRoaches, where some AI agents\nhave achieved the performance level of human professional players. However, for\nfull games, the current AI agents are still far from achieving human\nprofessional level performance. To bridge this gap, we present two full game AI\nagents in this paper - the AI agent TStarBot1 is based on deep reinforcement\nlearning over a flat action structure, and the AI agent TStarBot2 is based on\nhard-coded rules over a hierarchical action structure. Both TStarBot1 and\nTStarBot2 are able to defeat the built-in AI agents from level 1 to level 10 in\na full game (1v1 Zerg-vs-Zerg game on the AbyssalReef map), noting that level\n8, level 9, and level 10 are cheating agents with unfair advantages such as\nfull vision on the whole map and resource harvest boosting. To the best of our\nknowledge, this is the first public work to investigate AI agents that can\ndefeat the built-in AI in the StarCraft II full game.","url_abs":"http://arxiv.org/abs/1809.07193v3","url_pdf":"http://arxiv.org/pdf/1809.07193v3.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":"tstarbots-defeating-the-cheating-level","repo_url":"https://github.com/Tencent/TStarBots","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"tstarbots-defeating-the-cheating-level","repo_url":"https://github.com/ericborn/binarybot","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"ai-agent","task_name":"AI Agent"},{"task_slug":"decision-making","task_name":"Decision Making"},{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"real-time-strategy-games","task_name":"Real-Time Strategy Games"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"starcraft","task_name":"Starcraft"},{"task_slug":"starcraft-ii","task_name":"Starcraft II"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.07193","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}