5 citations · 7 across the 5 of their papers we have counts for
6 papers
Fighting Uncertainty with Gradients: Offline Reinforcement Learning via Diffusion Score Matching
H. J. Terry Suh, Glen Chou, Hongkai Dai +3
Gradient-based methods enable efficient search capabilities in high dimensions. However, in order to apply them effectively in offline optimization paradigms such as offline Reinfo…
Global Planning for Contact-Rich Manipulation via Local Smoothing of Quasi-dynamic Contact Models
Tao Pang, H. J. Terry Suh, Lujie Yang +1
The empirical success of Reinforcement Learning (RL) in the setting of contact-rich manipulation leaves much to be understood from a model-based perspective, where the key difficul…
Do Differentiable Simulators Give Better Policy Gradients?
H. J. Terry Suh, Max Simchowitz, Kaiqing Zhang +1
Differentiable simulators promise faster computation time for reinforcement learning by replacing zeroth-order gradient estimates of a stochastic objective with an estimate based o…
SEED: Series Elastic End Effectors in 6D for Visuotactile Tool Use
H. J. Terry Suh, Naveen Kuppuswamy, Tao Pang +3
We propose the framework of Series Elastic End Effectors in 6D (SEED), which combines a spatially compliant element with visuotactile sensing to grasp and manipulate tools in the w…
The Surprising Effectiveness of Linear Models for Visual Foresight in Object Pile Manipulation
H. J. Terry Suh, Russ Tedrake
In this paper, we tackle the problem of pushing piles of small objects into a desired target set using visual feedback. Unlike conventional single-object manipulation pipelines, wh…
Energy-Efficient Motion Planning for Multi-Modal Hybrid Locomotion
H. J. Terry Suh, Xiaobin Xiong, Andrew Singletary +2
Hybrid locomotion, which combines multiple modalities of locomotion within a single robot, enables robots to carry out complex tasks in diverse environments. This paper presents a…