{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T00:03:42Z","timestamp":1783123422510,"version":"3.54.6"},"reference-count":38,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,5,13]],"date-time":"2024-05-13T00:00:00Z","timestamp":1715558400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,5,13]],"date-time":"2024-05-13T00:00:00Z","timestamp":1715558400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,5,13]]},"DOI":"10.1109\/icra57147.2024.10610919","type":"proceedings-article","created":{"date-parts":[[2024,8,8]],"date-time":"2024-08-08T17:51:05Z","timestamp":1723139465000},"page":"12942-12948","source":"Crossref","is-referenced-by-count":3,"title":["Merging Decision Transformers: Weight Averaging for Forming Multi-Task Policies"],"prefix":"10.1109","author":[{"given":"Daniel","family":"Lawson","sequence":"first","affiliation":[{"name":"Purdue University,Department of Computer Science,West Lafayette,IN,USA,47907"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ahmed H.","family":"Qureshi","sequence":"additional","affiliation":[{"name":"Purdue University,Department of Computer Science,West Lafayette,IN,USA,47907"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","author":"Ainsworth","year":"2022","journal-title":"Git re-basin: Merging models modulo permutation symmetries"},{"key":"ref2","author":"Ba","year":"2016","journal-title":"Layer normalization"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/iros55552.2023.10342381"},{"key":"ref4","author":"Chen","year":"2021","journal-title":"Decision transformer: Reinforcement learning via sequence modeling"},{"key":"ref5","author":"Choshen","year":"2022","journal-title":"Fusing finetuned models for better pretraining"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-eacl.153"},{"key":"ref7","author":"Don-Yehiya","year":"2022","journal-title":"Cold fusion: Collaborative descent for distributed multitask finetuning"},{"key":"ref8","author":"Du","year":"2023","journal-title":"Learning universal policies via text-guided video generation"},{"key":"ref9","author":"Entezari","year":"2021","journal-title":"The role of permutation invariance in linear mode connectivity of neural networks"},{"key":"ref10","author":"Fu","year":"2020","journal-title":"D4rl: Datasets for deep data-driven reinforcement learning"},{"key":"ref11","author":"Furuta","year":"2021","journal-title":"Generalized decision transformer for offline hindsight information matching"},{"key":"ref12","author":"Ilharco","year":"2022","journal-title":"Editing models with task arithmetic"},{"key":"ref13","author":"Ilharco","year":"2022","journal-title":"Patching open-vocabulary models by interpolating weights"},{"key":"ref14","author":"Jiang","year":"2022","journal-title":"Vima-manipulation"},{"key":"ref15","author":"Jin","year":"2022","journal-title":"Dataless knowledge fusion by merging weights of language models"},{"key":"ref16","article-title":"Repair: Renormalizing permuted activations for interpolation repair","author":"Jordan","year":"2022"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1613\/jair.1.13673"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1611835114"},{"key":"ref19","author":"Kumar","year":"2022","journal-title":"Offline q-learning on diverse multi-task data both scales and generalizes"},{"key":"ref20","author":"Kumar","year":"2022","journal-title":"Pre-training for robots: Offline rl enables learning new tasks from a handful of trials"},{"key":"ref21","author":"Lee","year":"2022","journal-title":"Multi-game decision transformers"},{"key":"ref22","author":"Li","year":"2022","journal-title":"Branch-train-merge: Embarrassingly parallel training of expert language models"},{"key":"ref23","author":"Lu","year":"2021","journal-title":"Pre-trained transformers as universal computation engines"},{"key":"ref24","author":"Matena","year":"2021","journal-title":"Merging models with fisher-weighted averaging"},{"key":"ref25","author":"McMahan","year":"2016","journal-title":"Communication-efficient learning of deep networks from decentralized data"},{"key":"ref26","author":"Merity","year":"2016","journal-title":"Pointer sentinel mixture models"},{"key":"ref27","author":"Guerrero Pe\u00f1a","year":"2022","journal-title":"Re-basin via implicit sinkhorn differentiation"},{"key":"ref28","author":"Phuong","year":"2022","journal-title":"Formal algorithms for transformers"},{"key":"ref29","author":"Radford","year":"2018","journal-title":"Improving language understanding by generative pre-training"},{"key":"ref30","author":"Reed","year":"2022","journal-title":"A generalist agent"},{"key":"ref31","author":"Reid","year":"2022","journal-title":"Can wikipedia help offline reinforcement learning?"},{"key":"ref32","article-title":"Investigating multi-task pretraining and generalization in reinforcement learning","volume-title":"The Eleventh International Conference on Learning Representations","author":"Taiga"},{"key":"ref33","author":"Takagi","year":"2022","journal-title":"On the effect of pre-training for transformer in different modality on offline reinforcement learning"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2012.6386109"},{"key":"ref35","author":"Vaswani","year":"2017","journal-title":"Attention is all you need"},{"key":"ref36","author":"Wortsman","year":"2022","journal-title":"Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52688.2022.00780"},{"key":"ref38","author":"Xu","year":"2022","journal-title":"On the feasibility of cross-task transfer with model-based reinforcement learning"}],"event":{"name":"2024 IEEE International Conference on Robotics and Automation (ICRA)","location":"Yokohama, Japan","start":{"date-parts":[[2024,5,13]]},"end":{"date-parts":[[2024,5,17]]}},"container-title":["2024 IEEE International Conference on Robotics and Automation (ICRA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10609961\/10609862\/10610919.pdf?arnumber=10610919","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,10]],"date-time":"2024-08-10T05:59:13Z","timestamp":1723269553000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10610919\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,5,13]]},"references-count":38,"URL":"https:\/\/doi.org\/10.1109\/icra57147.2024.10610919","relation":{},"subject":[],"published":{"date-parts":[[2024,5,13]]}}}