ziyu wang

Papers

Total Citations

H-Index

About

Ziyu Wang is a leading researcher in reinforcement learning, with a particular focus on off-policy evaluation (OPE) and offline decision-making. Their most cited work, "Benchmarks for Deep Off-Policy Evaluation" (2021, 25 citations), addresses a critical challenge in deploying RL in high-stakes domains like healthcare and recommender systems: how to reliably evaluate and select policies using only pre-collected, static datasets. By establishing standardized benchmarks, Wang has provided the research community with essential tools for comparing OPE methods, directly enabling safer and more robust deployment of AI in settings where online experimentation is impractical or dangerous. This contribution is foundational for advancing offline RL, a field that seeks to learn effective decision-making strategies without costly or risky real-world interactions. Wang's work bridges the gap between theoretical OPE methods and practical applications, making them a key figure in the push toward reliable, data-driven policy evaluation. Their research continues to shape how practitioners assess and trust learned policies, with growing influence across both academic and applied AI communities.