Recently, people have shown that large-scale pre-training from internet-scale data is the key to building generalist models, as witnessed in NLP. To build embodied generalist agents …
Video instance segmentation requires classifying segmenting and tracking every object across video frames. Unlike existing approaches that rely on masks boxes or category labels …
Dyna-style model-based reinforcement learning contains two phases: model rollouts to generate sample for policy learning and real environment exploration using current policy …
Learning rewards from expert videos offers an affordable and effective solution to specify the intended behaviors for reinforcement learning tasks. In this work, we propose Diffusion …
We present Premier-TACO, a multitask feature representation learning approach designed to improve few-shot policy learning efficiency in sequential decision-making tasks. Premier …
Recently, various pre-training methods have been introduced in vision-based Reinforcement Learning (RL). However, their generalization ability remains unclear due to …
Z Yuan, T Wei, S Cheng, G Zhang, Y Chen… - arXiv preprint arXiv …, 2024 - arxiv.org
Can we endow visuomotor robots with generalization capabilities to operate in diverse open- world scenarios? In this paper, we propose\textbf {Maniwhere}, a generalizable framework …
A Mete, H Xue, A Wilcox, Y Chen, A Garg - arXiv preprint arXiv …, 2024 - arxiv.org
Generalization capabilities, or rather a lack thereof, is one of the most important unsolved problems in the field of robot learning, and while several large scale efforts have set out to …
Learning representations for reinforcement learning (RL) has shown much promise for continuous control. We propose an efficient representation learning method using only a self …