activity
20212024
most citedEfficient Adversarial Training without Attacking: Worst-Case-Aware Robust Reinforcement Learning

8 citations · 27 across the 15 of their papers we have counts for

collaborators

16 papers

cs.RO2024★ 1 cited

TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policies

Ruijie Zheng, Yongyuan Liang, Shuaiyi Huang +5

Although large vision-language-action (VLA) models pretrained on extensive robot datasets offer promising generalist policies for robotic learning, they still struggle with spatial…

cs.LG2024★ 2 cited

ACE : Off-Policy Actor-Critic with Causality-Aware Entropy Regularization

Tianying Ji, Yongyuan Liang, Yan Zeng +7

The varying significance of distinct primitive behaviors during the policy learning process has been overlooked by prior model-free RL algorithms. Leveraging this insight, we explo…

cs.LG2024

PRISE: LLM-Style Sequence Compression for Learning Temporal Action Abstractions in Control

Ruijie Zheng, Ching-An Cheng, Hal Daumé +2

Temporal action abstractions, along with belief state representations, are a powerful knowledge sharing mechanism for sequential decision making. In this work, we propose a novel v…

cs.LG2024

Premier-TACO is a Few-Shot Policy Learner: Pretraining Multitask Representation via Temporal Action-Driven Contrastive Loss

Ruijie Zheng, Yongyuan Liang, Xiyao Wang +7

We present Premier-TACO, a multitask feature representation learning approach designed to improve few-shot policy learning efficiency in sequential decision-making tasks. Premier-T…

cs.LG2023

Progressively Efficient Learning

Ruijie Zheng, Khanh Nguyen, Hal Daumé +2

Assistant AI agents should be capable of rapidly acquiring novel skills and adapting to new user preferences. Traditional frameworks like imitation learning and reinforcement learn…

cs.LG2023★ 3 cited

DrM: Mastering Visual Reinforcement Learning through Dormant Ratio Minimization

Guowei Xu, Ruijie Zheng, Yongyuan Liang +12

Visual reinforcement learning (RL) has shown promise in continuous control tasks. Despite its progress, current algorithms are still unsatisfactory in virtually every aspect of the…