{"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/distributed-distributional-deterministic","title":"Distributed Distributional Deterministic Policy Gradients","arxiv_id":"1804.08617","date":"2018-04-23","proceeding":"ICLR 2018 1","authors":["Gabriel Barth-Maron","Matthew W. Hoffman","David Budden","Will Dabney","Dan Horgan","Dhruva TB","Alistair Muldal","Nicolas Heess","Timothy Lillicrap"],"abstract":"This work adopts the very successful distributional perspective on\nreinforcement learning and adapts it to the continuous control setting. We\ncombine this within a distributed framework for off-policy learning in order to\ndevelop what we call the Distributed Distributional Deep Deterministic Policy\nGradient algorithm, D4PG. We also combine this technique with a number of\nadditional, simple improvements such as the use of $N$-step returns and\nprioritized experience replay. Experimentally we examine the contribution of\neach of these individual components, and show how they interact, as well as\ntheir combined contributions. Our results show that across a wide variety of\nsimple control tasks, difficult manipulation tasks, and a set of hard\nobstacle-based locomotion tasks the D4PG algorithm achieves state of the art\nperformance.","url_abs":"http://arxiv.org/abs/1804.08617v1","url_pdf":"http://arxiv.org/pdf/1804.08617v1.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":"distributed-distributional-deterministic","repo_url":"https://github.com/schatty/D4PG-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"distributed-distributional-deterministic","repo_url":"https://github.com/vgudapati/DRLND_Collaboration_Competetion","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"distributed-distributional-deterministic","repo_url":"https://github.com/zhou-henry/distributed-distributional-drq","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"distributed-distributional-deterministic","repo_url":"https://github.com/HzcIrving/DLRL-PlayGround","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"distributed-distributional-deterministic","repo_url":"https://github.com/opendilab/DI-engine/blob/main/ding/policy/d4pg.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"continuous-control","task_name":"Continuous Control"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"continuous-control","task_name":"continuous-control"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"d4pg","method_name":"D4PG"},{"method_slug":"n-step-returns","method_name":"N-step Returns"},{"method_slug":"prioritized-experience-replay","method_name":"Prioritized Experience Replay"}],"datasets_introduced":[],"methods_introduced":[{"slug":"d4pg","name":"D4PG","full_name":"Distributed Distributional DDPG"}],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.08617","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}