Papers
2
Total Citations
9
H-Index
2
About
Ziyang Tang is a researcher at the intersection of reinforcement learning and robotics, with key contributions in off-policy evaluation and humanoid locomotion. His most cited work, "Accountable Off-Policy Evaluation With Kernel Bellman Statistics" (2020, 7 citations), introduces a statistically rigorous framework for assessing new policies from historical data—critical for high-stakes domains like healthcare where direct policy execution is risky. This work advances trustworthy decision-making under uncertainty by leveraging kernel methods to provide finite-sample guarantees. More recently, Tang’s 2025 paper "Think on Your Feet" (2 citations) tackles the challenge of enabling humanoid robots to fluidly transition between diverse locomotion skills in response to real-time commands, blending motion imitation with adaptive control. This work pushes toward more natural, responsive human-robot interaction. Tang’s research bridges theoretical rigor and practical deployment, with his OPE methods offering foundational tools for safe policy evaluation, while his robotics work demonstrates a commitment to embodied intelligence. His growing citation record reflects the relevance of his contributions to both algorithmic accountability and agile robotic systems.
Research Focus
Key Achievements
Top Papers
- 1Accountable Off-Policy Evaluation With Kernel Bellman Statistics7 citations · 2020
- 2