{"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/alchemy-a-structured-task-distribution-for","title":"Alchemy: A benchmark and analysis toolkit for meta-reinforcement learning agents","arxiv_id":"2102.02926","date":"2021-02-04","proceeding":null,"authors":["Jane X. Wang","Michael King","Nicolas Porcel","Zeb Kurth-Nelson","Tina Zhu","Charlie Deck","Peter Choy","Mary Cassin","Malcolm Reynolds","Francis Song","Gavin Buttimore","David P. Reichert","Neil Rabinowitz","Loic Matthey","Demis Hassabis","Alexander Lerchner","Matthew Botvinick"],"abstract":"There has been rapidly growing interest in meta-learning as a method for increasing the flexibility and sample efficiency of reinforcement learning. One problem in this area of research, however, has been a scarcity of adequate benchmark tasks. In general, the structure underlying past benchmarks has either been too simple to be inherently interesting, or too ill-defined to support principled analysis. In the present work, we introduce a new benchmark for meta-RL research, emphasizing transparency and potential for in-depth analysis as well as structural richness. Alchemy is a 3D video game, implemented in Unity, which involves a latent causal structure that is resampled procedurally from episode to episode, affording structure learning, online inference, hypothesis testing and action sequencing based on abstract domain knowledge. We evaluate a pair of powerful RL agents on Alchemy and present an in-depth analysis of one of these agents. Results clearly indicate a frank and specific failure of meta-learning, providing validation for Alchemy as a challenging benchmark for meta-RL. Concurrent with this report, we are releasing Alchemy as public resource, together with a suite of analysis tools and sample agent trajectories.","url_abs":"https://arxiv.org/abs/2102.02926v3","url_pdf":"https://arxiv.org/pdf/2102.02926v3.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":"alchemy-a-structured-task-distribution-for","repo_url":"https://github.com/deepmind/dm_alchemy","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"meta-reinforcement-learning","task_name":"Meta Reinforcement Learning"},{"task_slug":"meta-learning","task_name":"Meta-Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"unity","task_name":"Unity"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[{"slug":"alchemy","name":"Alchemy","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2102.02926","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}