{"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/one-shot-imitation-from-observing-humans-via","title":"One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning","arxiv_id":"1802.01557","date":"2018-02-05","proceeding":null,"authors":["Tianhe Yu","Chelsea Finn","Annie Xie","Sudeep Dasari","Tianhao Zhang","Pieter Abbeel","Sergey Levine"],"abstract":"Humans and animals are capable of learning a new behavior by observing others\nperform the skill just once. We consider the problem of allowing a robot to do\nthe same -- learning from a raw video pixels of a human, even when there is\nsubstantial domain shift in the perspective, environment, and embodiment\nbetween the robot and the observed human. Prior approaches to this problem have\nhand-specified how human and robot actions correspond and often relied on\nexplicit human pose detection systems. In this work, we present an approach for\none-shot learning from a video of a human by using human and robot\ndemonstration data from a variety of previous tasks to build up prior knowledge\nthrough meta-learning. Then, combining this prior knowledge and only a single\nvideo demonstration from a human, the robot can perform the task that the human\ndemonstrated. We show experiments on both a PR2 arm and a Sawyer arm,\ndemonstrating that after meta-learning, the robot can learn to place, push, and\npick-and-place new objects using just one video of a human performing the\nmanipulation.","url_abs":"http://arxiv.org/abs/1802.01557v1","url_pdf":"http://arxiv.org/pdf/1802.01557v1.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":"one-shot-imitation-from-observing-humans-via","repo_url":"https://github.com/daiyk/daml_pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"one-shot-imitation-from-observing-humans-via","repo_url":"https://github.com/tianheyu927/mil","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"meta-learning","task_name":"Meta-Learning"},{"task_slug":"one-shot-learning","task_name":"One-Shot Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1802.01557","atlas_url":"https://app.syntology.ai/?focus=1802.01557","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}