{"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/starcraft-micromanagement-with-reinforcement","title":"StarCraft Micromanagement with Reinforcement Learning and Curriculum Transfer Learning","arxiv_id":"1804.00810","date":"2018-04-03","proceeding":null,"authors":["Kun Shao","Yuanheng Zhu","Dongbin Zhao"],"abstract":"Real-time strategy games have been an important field of game artificial\nintelligence in recent years. This paper presents a reinforcement learning and\ncurriculum transfer learning method to control multiple units in StarCraft\nmicromanagement. We define an efficient state representation, which breaks down\nthe complexity caused by the large state space in the game environment. Then a\nparameter sharing multi-agent gradientdescent Sarsa({\\lambda}) (PS-MAGDS)\nalgorithm is proposed to train the units. The learning policy is shared among\nour units to encourage cooperative behaviors. We use a neural network as a\nfunction approximator to estimate the action-value function, and propose a\nreward function to help units balance their move and attack. In addition, a\ntransfer learning method is used to extend our model to more difficult\nscenarios, which accelerates the training process and improves the learning\nperformance. In small scale scenarios, our units successfully learn to combat\nand defeat the built-in AI with 100% win rates. In large scale scenarios,\ncurriculum transfer learning method is used to progressively train a group of\nunits, and shows superior performance over some baseline methods in target\nscenarios. With reinforcement learning and curriculum transfer learning, our\nunits are able to learn appropriate strategies in StarCraft micromanagement\nscenarios.","url_abs":"http://arxiv.org/abs/1804.00810v1","url_pdf":"http://arxiv.org/pdf/1804.00810v1.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":"starcraft-micromanagement-with-reinforcement","repo_url":"https://github.com/nanxintin/StarCraft-AI","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"real-time-strategy-games","task_name":"Real-Time Strategy Games"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"starcraft","task_name":"Starcraft"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1804.00810","atlas_url":"https://app.syntology.ai/?focus=1804.00810","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}