Zobrazeno 1 - 10 of 198 pro vyhledávání: '"Thomas, Garrett A."'
1
Report
PLEX: Making the Most of the Available Data for Robotic Manipulation Pretraining
Autor: Thomas, Garrett, Cheng, Ching-An, Loynd, Ricky, Frujeri, Felipe Vieira, Vineet, Vibhav, Jalobeanu, Mihai, Kolobov, Andrey
A rich representation is key to general robotic manipulation, but existing approaches to representation learning require large amounts of multimodal demonstrations. In this work we propose PLEX, a transformer-based architecture that learns from a sma
Externí odkaz: http://arxiv.org/abs/2303.08789
Zobrazit plný text záznamu
2
Report
Safe Reinforcement Learning by Imagining the Near Future
Autor: Thomas, Garrett, Luo, Yuping, Ma, Tengyu
Safe reinforcement learning is a promising path toward applying reinforcement learning algorithms to real-world problems, where suboptimal behaviors may lead to actual negative consequences. In this work, we focus on the setting where unsafe states c
Externí odkaz: http://arxiv.org/abs/2202.07789
Zobrazit plný text záznamu
7
Report
Model-based Adversarial Meta-Reinforcement Learning
Autor: Lin, Zichuan, Thomas, Garrett, Yang, Guangwen, Ma, Tengyu
Meta-reinforcement learning (meta-RL) aims to learn from multiple training tasks the ability to adapt efficiently to unseen test tasks. Despite the success, existing meta-RL algorithms are known to be sensitive to the task distribution shift. When th
Externí odkaz: http://arxiv.org/abs/2006.08875
Zobrazit plný text záznamu
8
Report
MOPO: Model-based Offline Policy Optimization
Autor: Yu, Tianhe, Thomas, Garrett, Yu, Lantao, Ermon, Stefano, Zou, James, Levine, Sergey, Finn, Chelsea, Ma, Tengyu
Offline reinforcement learning (RL) refers to the problem of learning policies entirely from a large batch of previously collected data. This problem setting offers the promise of utilizing such datasets to acquire policies without any costly or dang
Externí odkaz: http://arxiv.org/abs/2005.13239
Zobrazit plný text záznamu
9
Report
A Model-based Approach for Sample-efficient Multi-task Reinforcement Learning
Autor: Landolfi, Nicholas C., Thomas, Garrett, Ma, Tengyu
The aim of multi-task reinforcement learning is two-fold: (1) efficiently learn by training against multiple tasks and (2) quickly adapt, using limited samples, to a variety of new tasks. In this work, the tasks correspond to reward functions for env
Externí odkaz: http://arxiv.org/abs/1907.04964
Zobrazit plný text záznamu
10
Report
Learning Robotic Assembly from CAD
Autor: Thomas, Garrett, Chien, Melissa, Tamar, Aviv, Ojea, Juan Aparicio, Abbeel, Pieter
In this work, motivated by recent manufacturing trends, we investigate autonomous robotic assembly. Industrial assembly tasks require contact-rich manipulation skills, which are challenging to acquire using classical control and motion planning appro
Externí odkaz: http://arxiv.org/abs/1803.07635
Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání

Omezení vyhledávání
Plný text Recenzováno Digitální knihovna AV ČR
Zdroje