{"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/efficient-parallel-methods-for-deep","title":"Efficient Parallel Methods for Deep Reinforcement Learning","arxiv_id":"1705.04862","date":"2017-05-13","proceeding":null,"authors":["Alfredo V. Clemente","Humberto N. Castejón","Arjun Chandra"],"abstract":"We propose a novel framework for efficient parallelization of deep\nreinforcement learning algorithms, enabling these algorithms to learn from\nmultiple actors on a single machine. The framework is algorithm agnostic and\ncan be applied to on-policy, off-policy, value based and policy gradient based\nalgorithms. Given its inherent parallelism, the framework can be efficiently\nimplemented on a GPU, allowing the usage of powerful models while significantly\nreducing training time. We demonstrate the effectiveness of our framework by\nimplementing an advantage actor-critic algorithm on a GPU, using on-policy\nexperiences and employing synchronous updates. Our algorithm achieves\nstate-of-the-art performance on the Atari domain after only a few hours of\ntraining. Our framework thus opens the door for much faster experimentation on\ndemanding problem domains. Our implementation is open-source and is made public\nat https://github.com/alfredvc/paac","url_abs":"http://arxiv.org/abs/1705.04862v2","url_pdf":"http://arxiv.org/pdf/1705.04862v2.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":"efficient-parallel-methods-for-deep","repo_url":"https://github.com/alfredvc/paac","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"efficient-parallel-methods-for-deep","repo_url":"https://github.com/jcwleo/curiosity-driven-exploration-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"efficient-parallel-methods-for-deep","repo_url":"https://github.com/jcwleo/random-network-distillation-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"efficient-parallel-methods-for-deep","repo_url":"https://github.com/learnermaxRL/PPO_A2C","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"efficient-parallel-methods-for-deep","repo_url":"https://github.com/pkiourti/rl_backdoor","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1705.04862","atlas_url":"https://app.syntology.ai/?focus=1705.04862","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}