{"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/learning-complex-dexterous-manipulation-with","title":"Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations","arxiv_id":"1709.10087","date":"2017-09-28","proceeding":null,"authors":["Aravind Rajeswaran","Vikash Kumar","Abhishek Gupta","Giulia Vezzani","John Schulman","Emanuel Todorov","Sergey Levine"],"abstract":"Dexterous multi-fingered hands are extremely versatile and provide a generic\nway to perform a multitude of tasks in human-centric environments. However,\neffectively controlling them remains challenging due to their high\ndimensionality and large number of potential contacts. Deep reinforcement\nlearning (DRL) provides a model-agnostic approach to control complex dynamical\nsystems, but has not been shown to scale to high-dimensional dexterous\nmanipulation. Furthermore, deployment of DRL on physical systems remains\nchallenging due to sample inefficiency. Consequently, the success of DRL in\nrobotics has thus far been limited to simpler manipulators and tasks. In this\nwork, we show that model-free DRL can effectively scale up to complex\nmanipulation tasks with a high-dimensional 24-DoF hand, and solve them from\nscratch in simulated experiments. Furthermore, with the use of a small number\nof human demonstrations, the sample complexity can be significantly reduced,\nwhich enables learning with sample sizes equivalent to a few hours of robot\nexperience. The use of demonstrations result in policies that exhibit very\nnatural movements and, surprisingly, are also substantially more robust.","url_abs":"http://arxiv.org/abs/1709.10087v2","url_pdf":"http://arxiv.org/pdf/1709.10087v2.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":"learning-complex-dexterous-manipulation-with","repo_url":"https://github.com/avivne/bilinear-transduction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"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":{"atlas_url":"https://app.syntology.ai/?focus=1709.10087","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}