Haoran Tang
Papers
1
Total Citations
434
H-Index
1
About
Haoran Tang is a leading researcher in reinforcement learning, whose work has fundamentally advanced the field of energy-based and maximum entropy policies. His most influential contribution is the seminal paper "Reinforcement Learning with Deep Energy-Based Policies" (2017, 434 citations), where he introduced soft Q-learning—a groundbreaking algorithm that made it feasible to learn expressive energy-based policies for continuous states and actions, a challenge previously confined to tabular domains. This work not only provided a rigorous framework for maximum entropy reinforcement learning but also demonstrated how to derive optimal stochastic policies that balance exploration and exploitation. Tang's research has had a profound impact on the development of modern RL algorithms, influencing subsequent work on soft actor-critic and entropy-regularized methods. His ability to bridge theoretical foundations with practical algorithms has made his contributions essential reading for anyone working in deep reinforcement learning, robotics, and decision-making under uncertainty.
Research Focus
Key Achievements
Top Papers
- 1Reinforcement Learning with Deep Energy-Based Policies434 citations · 2017