10 papers
Cognitive Action Reasoning for Proactive Robots from Human-Centered Multimodal Observations
Zhihao Gu, Kechao Zhu, Yuanfeng Wu +7
Robots operating in human-centered environments are typically designed to execute explicit instructions, and most robot-learning datasets likewise pair observations with task instr…
An Lower Bound for Stochastic NC-SC Bilevel Optimization with First-order Oracles
Zhihao Gu, Qilong Wu, Junchi Yang
We study the oracle complexity of finding -stationary points of smooth bilevel optimization problems with a nonconvex upper-level objective and a strongly convex lower-level pro…
Lower Bounds for Nonconvex-Concave Minimax Optimization
Qilong Wu, Zhihao Gu, Junchi Yang
We study lower bounds on the first-order oracle complexity of smooth nonconvex-concave minimax optimization. We consider objectives that are jointly -smooth in the primal an…
SGHA: A Single-Loop Fully First-Order Algorithm for Nonconvex-Strongly-Convex Bilevel Optimization
Zhihao Gu, Qilong Wu, Junchi Yang
In this work, we study the oracle complexity of finding an -stationary point for nonconvex-strongly-convex (NC-SC) bilevel optimization using only first-order oracles. Existing…
VistaVLA: Geometry- and Semantic-Aware 3D Gaussian-Grounded VLA for Robotic Manipulation
Mohan Liu, Zhihao Gu, Xuanyu Chen +5
Vision-Language-Action (VLA) models have emerged as a powerful end-to-end paradigm for robotic manipulation by mapping language instructions and 2D visual inputs directly to action…
Physics-Guided Biomechanical Gait Adaptation for Humanoid Locomotion on Extreme Sloped Terrains
Xuanyu Chen, Mohan Liu, Dengchen Mei +6
Model-free reinforcement learning has enabled impressive humanoid locomotion; however, control on steep slopes remains largely unexplored. Unlike flat or discrete terrains, sloped…
LiMoDE: Rethinking Lifelong Robot Manipulation from a Mixture-of-Dynamic-Experts Perspective
Zhihao Gu, Lin Wang
Building a generalist robot that can leverage prior knowledge for continuous task adaptation remains a significant challenge. Previous works alleviate the catastrophic forgetting p…
Physics-Informed Policy Optimization via Analytic Dynamics Regularization
Namai Chandra, Liu Mohan, Zhihao Gu +1
Reinforcement learning (RL) has achieved strong performance in robotic control; however, state-of-the-art policy learning methods, such as actor-critic methods, still suffer from h…
Learning Diffusion Policy from Primitive Skills for Robot Manipulation
Zhihao Gu, Ming Yang, Difan Zou +1
Diffusion policies (DP) have recently shown great promise for generating actions in robotic manipulation. However, existing approaches often rely on global instructions to produce…
Zeroth-Order Stochastic Mirror Descent Algorithms for Minimax Excess Risk Optimization
Zhihao Gu, Zi Xu
The minimax excess risk optimization (MERO) problem is a new variation of the traditional distributionally robust optimization (DRO) problem, which achieves uniformly low regret ac…