Hao Liang




Causality Meets Locality: Provably Generalizable and Scalable Policy Learning for Networked Systems [link] NeurIPS 2025 (Spotlight)
Hao Liang*, Shuqing Shi*, Yudi Zhang, Biwei Huang, Yali Du

Why GRPO Needs Normalization: A Local-Curvature Perspective on Adaptive Gradients To appear
Cheng Ge*, Heqi Yin*, Hao Liang, Jiawei Zhang
* Equal contribution. † Co-last/corresponding authors.
Scheduled for poster presentation at NeurIPS 2025 Workshop on Efficient Reasoning.

Is Pure Exploitation Sufficient for Sequential Decision-Making with Exogenous Information? To appear
Hao Liang, Jiayu Cheng, Sean R. Sinclair, Yali Du


Bridging Distributional and Risk-sensitive Reinforcement Learning with Provable Regret Bounds [link] JMLR
Hao Liang, Zhi-Quan Luo
Scheduled for poster presentation at NeurIPS 2025 (The NeurIPS/ICLR/ICML Journal-to-Conference Track). An earlier version presented at NeurIPS 2021 Workshop on Ecological Theory of RL

Bridging Distributional and Risk-Sensitive Reinforcement Learning: Balancing Statistical, Computational, and Risk Considerations [link]
Hao Liang
Present at ICML 2024 Workshop on FoRLaC

Regret Bounds for Risk-sensitive Reinforcement Learning with Lipschitz Dynamic Risk Measures [AISTATS] AISTATS 2024
Hao Liang, Zhi-Quan Luo
An earlier version presented at ICML 2023 Workshop on New Frontiers in Learning, Control, and Dynamical Systems

A Distribution Optimization Framework for Confidence Bounds of Risk Measures [ICML] ICML 2023
Hao Liang, Zhi-Quan Luo


Working Papers


Revisiting Minimax Lower Bounds in Unknown Matrix Game
Hao Liang, Yingru Li