Liming Xiao
Papers
1
Total Citations
8
H-Index
1
About
Liming Xiao is an emerging researcher specializing in deep reinforcement learning, with a particular focus on improving the reliability and efficiency of model-free algorithms for complex decision-making and control tasks. Their most notable work, "Distributional Soft Actor-Critic with Three Refinements" (2023), addresses one of the field's persistent challenges: inaccurate value estimation, especially the overestimation of Q-values that can derail policy learning and lead to suboptimal performance. By integrating distributional reinforcement learning principles with targeted algorithmic refinements into the Soft Actor-Critic framework, Xiao's research pushes toward more stable and accurate policy optimization. Though still an early-career contributor with 8 citations on this paper, the work tackles a foundational problem that has broad implications for robotics, autonomous systems, and game-playing agents. Xiao's contributions represent a meaningful step forward in making model-free RL algorithms more robust and practically deployable, positioning them as a promising voice in the ongoing effort to close the gap between theoretical promise and real-world performance in reinforcement learning research.
Research Focus
Key Achievements
Top Papers
- 1Distributional Soft Actor-Critic with Three Refinements8 citations · 2023