{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T18:24:13Z","timestamp":1779906253493,"version":"3.53.1"},"reference-count":24,"publisher":"IEEE","license":[{"start":{"date-parts":[[2021,9,27]],"date-time":"2021-09-27T00:00:00Z","timestamp":1632700800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2021,9,27]],"date-time":"2021-09-27T00:00:00Z","timestamp":1632700800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2021,9,27]],"date-time":"2021-09-27T00:00:00Z","timestamp":1632700800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100004721","name":"University of Tokyo","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100004721","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,9,27]]},"DOI":"10.1109\/iros51168.2021.9636301","type":"proceedings-article","created":{"date-parts":[[2021,12,16]],"date-time":"2021-12-16T20:45:38Z","timestamp":1639687538000},"page":"8965-8972","source":"Crossref","is-referenced-by-count":50,"title":["Transformer-based deep imitation learning for dual-arm robot manipulation"],"prefix":"10.1109","author":[{"given":"Heecheol","family":"Kim","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yoshiyuki","family":"Ohmura","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yasuo","family":"Kuniyoshi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref10","article-title":"Deep imitation learning for bimanual robotic manipulation","author":"xie","year":"2020","journal-title":"Neural Information Processing Systems"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2020.2998410"},{"key":"ref12","first-page":"5998","article-title":"Attention is all you need","author":"vaswani","year":"2017","journal-title":"Neural Information Processing Systems"},{"key":"ref13","author":"devlin","year":"2018","journal-title":"BERT Pre-training of deep bidirectional transformers for language understanding"},{"key":"ref14","author":"brown","year":"2020","journal-title":"Language models are few-shot learners"},{"key":"ref15","author":"dosovitskiy","year":"2020","journal-title":"An image is worth 16x16 words Transformers for image recognition at scale"},{"key":"ref16","first-page":"213","article-title":"End-to-end object detection with transformers","author":"carion","year":"2020","journal-title":"European Conference on Computer Vision"},{"key":"ref17","author":"esser","year":"2020","journal-title":"Taming transformers for highresolution image synthesis"},{"key":"ref18","author":"kaiser","year":"2017","journal-title":"One model to learn them all"},{"key":"ref19","article-title":"Transformers for one-shot visual imitation","author":"dasari","year":"2020","journal-title":"Conference on Robot Learning"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/j.robot.2012.07.005"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2021.3059619"},{"key":"ref6","first-page":"479","article-title":"Dualarm robotic manipulation of flexible cables","author":"zhu","year":"2018","journal-title":"International Conference on Intelligent Robots and Systems"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2015.7139991"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1142\/S0219843608001431"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.3006919"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2018.8461249"},{"key":"ref1","article-title":"Repeatable folding task by humanoid robot worker using deep learning","volume":"2","author":"yang","year":"2016","journal-title":"Robot and Automat Lett"},{"key":"ref9","first-page":"66","article-title":"Imitation learning and attentional supervision of dualarm structured tasks","author":"caccavale","year":"2017","journal-title":"International Conference on Development and Learning and Epigenetic Robotics"},{"key":"ref20","article-title":"Transformer-based meta-Imitation learning for robotic manipulation","author":"cachet","year":"2020","journal-title":"Neural Information Processing Systems Workshop on Robot Learning"},{"key":"ref22","author":"lin","year":"2013","journal-title":"Network in Network"},{"key":"ref21","article-title":"Mixture density networks","author":"bishop","year":"1994","journal-title":"Tech Rep"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.385"},{"key":"ref23","first-page":"1","article-title":"On the variance of the adaptive learning rate and beyond","author":"liu","year":"2020","journal-title":"International Conference on Learning Representations"}],"event":{"name":"2021 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","location":"Prague, Czech Republic","start":{"date-parts":[[2021,9,27]]},"end":{"date-parts":[[2021,10,1]]}},"container-title":["2021 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9635848\/9635849\/09636301.pdf?arnumber=9636301","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,10]],"date-time":"2022-05-10T16:54:37Z","timestamp":1652201677000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9636301\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9,27]]},"references-count":24,"URL":"https:\/\/doi.org\/10.1109\/iros51168.2021.9636301","relation":{},"subject":[],"published":{"date-parts":[[2021,9,27]]}}}