{"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/catalystrl-a-distributed-framework-for","title":"Catalyst.RL: A Distributed Framework for Reproducible RL Research","arxiv_id":"1903.00027","date":"2019-02-28","proceeding":null,"authors":["Sergey Kolesnikov","Oleksii Hrinchuk"],"abstract":"Despite the recent progress in deep reinforcement learning field (RL), and,\narguably because of it, a large body of work remains to be done in reproducing\nand carefully comparing different RL algorithms. We present catalyst.RL, an\nopen source framework for RL research with a focus on reproducibility and\nflexibility. Main features of our library include large-scale asynchronous\ndistributed training, easy-to-use configuration files with the complete list of\nhyperparameters for the particular experiments, efficient implementations of\nvarious RL algorithms and auxiliary tricks, such as frame stacking, n-step\nreturns, value distributions, etc. To vindicate the usefulness of our\nframework, we evaluate it on a range of benchmarks in a continuous control, as\nwell as on the task of developing a controller to enable a\nphysiologically-based human model with a prosthetic leg to walk and run. The\nlatter task was introduced at NeurIPS 2018 AI for Prosthetics Challenge, where\nour team took the 3rd place, capitalizing on the ability of catalyst.RL to\ntrain high-quality and sample-efficient RL agents.","url_abs":"http://arxiv.org/abs/1903.00027v1","url_pdf":"http://arxiv.org/pdf/1903.00027v1.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":"catalystrl-a-distributed-framework-for","repo_url":"https://github.com/catalyst-team/catalyst-rl-framework","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"continuous-control","task_name":"Continuous Control"},{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"continuous-control","task_name":"continuous-control"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1903.00027","atlas_url":"https://app.syntology.ai/?focus=1903.00027","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}