{"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/mazebase-a-sandbox-for-learning-from-games","title":"MazeBase: A Sandbox for Learning from Games","arxiv_id":"1511.07401","date":"2015-11-23","proceeding":null,"authors":["Sainbayar Sukhbaatar","Arthur Szlam","Gabriel Synnaeve","Soumith Chintala","Rob Fergus"],"abstract":"This paper introduces MazeBase: an environment for simple 2D games, designed\nas a sandbox for machine learning approaches to reasoning and planning. Within\nit, we create 10 simple games embodying a range of algorithmic tasks (e.g.\nif-then statements or set negation). A variety of neural models (fully\nconnected, convolutional network, memory network) are deployed via\nreinforcement learning on these games, with and without a procedurally\ngenerated curriculum. Despite the tasks' simplicity, the performance of the\nmodels is far from optimal, suggesting directions for future development. We\nalso demonstrate the versatility of MazeBase by using it to emulate small\ncombat scenarios from StarCraft. Models trained on the MazeBase version can be\ndirectly applied to StarCraft, where they consistently beat the in-game AI.","url_abs":"http://arxiv.org/abs/1511.07401v2","url_pdf":"http://arxiv.org/pdf/1511.07401v2.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":"mazebase-a-sandbox-for-learning-from-games","repo_url":"https://github.com/facebook/MazeBase","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"mazebase-a-sandbox-for-learning-from-games","repo_url":"https://github.com/facebookarchive/MazeBase","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"negation","task_name":"Negation"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"starcraft","task_name":"Starcraft"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}