{"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/deep-reinforcement-learning-for-vision-based","title":"Deep Reinforcement Learning for Vision-Based Robotic Grasping: A Simulated Comparative Evaluation of Off-Policy Methods","arxiv_id":"1802.10264","date":"2018-02-28","proceeding":null,"authors":["Deirdre Quillen","Eric Jang","Ofir Nachum","Chelsea Finn","Julian Ibarz","Sergey Levine"],"abstract":"In this paper, we explore deep reinforcement learning algorithms for\nvision-based robotic grasping. Model-free deep reinforcement learning (RL) has\nbeen successfully applied to a range of challenging environments, but the\nproliferation of algorithms makes it difficult to discern which particular\napproach would be best suited for a rich, diverse task like grasping. To answer\nthis question, we propose a simulated benchmark for robotic grasping that\nemphasizes off-policy learning and generalization to unseen objects. Off-policy\nlearning enables utilization of grasping data over a wide variety of objects,\nand diversity is important to enable the method to generalize to new objects\nthat were not seen during training. We evaluate the benchmark tasks against a\nvariety of Q-function estimation methods, a method previously proposed for\nrobotic grasping with deep neural network models, and a novel approach based on\na combination of Monte Carlo return estimation and an off-policy correction.\nOur results indicate that several simple methods provide a surprisingly strong\ncompetitor to popular algorithms such as double Q-learning, and our analysis of\nstability sheds light on the relative tradeoffs between the algorithms.","url_abs":"http://arxiv.org/abs/1802.10264v2","url_pdf":"http://arxiv.org/pdf/1802.10264v2.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":"deep-reinforcement-learning-for-vision-based","repo_url":"https://github.com/smrjan/robotic-vision","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"q-learning","task_name":"Q-Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"robotic-grasping","task_name":"Robotic Grasping"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1802.10264","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}