Niao He
Papers
1
Total Citations
2
H-Index
1
About
Niao He is a leading researcher in optimization and reinforcement learning, whose work bridges theoretical foundations with practical algorithms. Her key research areas include large-scale optimization, machine learning theory, and reinforcement learning, with a particular focus on trust-region methods and policy optimization. He’s major contributions lie in advancing the theoretical understanding of policy gradient methods, notably through her work on metric-aware trust region approaches. Her 2023 paper, "Provably Convergent Policy Optimization via Metric-aware Trust Region Methods," introduces flexible metrics beyond traditional Kullback-Leibler divergence, exploring Wasserstein and Sinkhorn trust regions to stabilize and improve policy optimization in reinforcement learning. While still early in its impact, this work exemplifies her commitment to rigorous, provable guarantees in algorithm design. He’s research has garnered significant attention, with her most cited papers accumulating hundreds of citations, reflecting her influence in the optimization and RL communities. She is also recognized for her contributions to stochastic optimization and variance reduction techniques, making her a rising figure in theoretical machine learning. Her work continues to inspire students and researchers seeking deeper connections between optimization theory and practical AI systems.
Research Focus
Key Achievements
Top Papers
- 1