Lixing Song

Rose–Hulman Institute of Technology

Papers

2

Total Citations

7

H-Index

2

About

Lixing Song is a researcher advancing the frontiers of reinforcement learning (RL) with a focus on data efficiency and safety. His work addresses two critical bottlenecks in modern RL: the need for massive trial data and the risk of unsafe exploration in real-world applications. In his 2021 paper, "A Data-Efficient Reinforcement Learning Method Based on Local Koopman Operators," Song introduced a model-based approach that leverages Koopman operator theory to dramatically reduce the data required for training, achieving strong asymptotic performance with fewer interactions—a contribution that has garnered 5 citations and offers a path toward practical RL deployment. Building on this, his 2022 work, "Safe Reinforcement Learning for LiDAR-based Navigation via Control Barrier Function," tackles safety head-on by integrating control barrier functions into RL policies for LiDAR-guided robots. This method ensures provably safe navigation even during trial-and-error learning, earning 2 citations and highlighting Song’s commitment to bridging theory and real-world safety constraints. Together, these contributions position Song as a key voice in making RL both efficient and trustworthy for autonomous systems.

Research Focus

Key Achievements

2
H-Index
2
Papers
7
Total Citations
4
Avg Citations/Paper
🏆 Most Cited Paper
A Data-Efficient Reinforcement Learning Method Based on Local Koopman Operators
5 citations · 2021
📈 Most Prolific Year: 2021 (1 Papers)
🤝 Key Collaborators: 4
🏛 Institutions: Rose–Hulman Institute of Technology

Top Papers

  1. 1
  2. 2

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago