no code implementations • 4 Mar 2024 • Toru Lin, Zhao-Heng Yin, Haozhi Qi, Pieter Abbeel, Jitendra Malik
Manipulating objects with two multi-fingered hands has been a long-standing challenge in robotics, attributed to the contact-rich nature of many manipulation tasks and the complexity inherent in coordinating a high-dimensional bimanual system.
no code implementations • 4 Dec 2023 • Ying Yuan, Haichuan Che, Yuzhe Qin, Binghao Huang, Zhao-Heng Yin, Kang-Won Lee, Yi Wu, Soo-Chul Lim, Xiaolong Wang
In this paper, we introduce a system that leverages visual and tactile sensory inputs to enable dexterous in-hand manipulation.
no code implementations • 11 Oct 2023 • Yuyang Liu, Weijun Dong, Yingdong Hu, Chuan Wen, Zhao-Heng Yin, Chongjie Zhang, Yang Gao
Nonetheless, we identify that tasks characterized by a progress dependency property pose significant challenges for such approaches; in these tasks, the agent needs to initially learn the expert's preceding behaviors before mastering the subsequent ones.
no code implementations • 23 Aug 2023 • John Ho, Zhao-Heng Yin, Colin Zhang, Nicole Guo, Yang Ha
Predicting the solubility of given molecules remains crucial in the pharmaceutical industry.
no code implementations • 20 Mar 2023 • Zhao-Heng Yin, Binghao Huang, Yuzhe Qin, Qifeng Chen, Xiaolong Wang
Relying on touch-only sensing, we can directly deploy the policy in a real robot hand and rotate novel objects that are not presented in training.
no code implementations • 17 Nov 2022 • Yuzhe Qin, Binghao Huang, Zhao-Heng Yin, Hao Su, Xiaolong Wang
We empirically evaluate our method using an Allegro Hand to grasp novel objects in both simulation and real world.
1 code implementation • 18 Oct 2022 • Zhao-Heng Yin, Weirui Ye, Qifeng Chen, Yang Gao
Inspired by the recent success of EfficientZero in RL, we propose EfficientImitate (EI), a planning-based imitation learning method that can achieve high in-environment sample efficiency and performance simultaneously.
1 code implementation • 13 Sep 2021 • Zhao-Heng Yin, Lingfeng Sun, Hengbo Ma, Masayoshi Tomizuka, Wu-Jun Li
In this paper, we consider CDIL on a class of similar robots.
no code implementations • 1 Mar 2021 • Zhao-Heng Yin, Lingfeng Sun, Liting Sun, Masayoshi Tomizuka, Wei Zhan
Experiments show that our model can generate diverse interactions in various scenarios.
no code implementations • 11 May 2020 • Zhao-Heng Yin, Wu-Jun Li
In particular, we propose planning to explore, in which TOMA is used to accelerate exploration by guiding the agent towards unexplored states.