Catherine Huang

University of California, Berkeley

Papers

1

Total Citations

8

H-Index

1

About

Catherine Huang is a leading researcher at the intersection of meta-reinforcement learning and offline reinforcement learning. Her most influential work, "Offline Meta-Reinforcement Learning with Online Self-Supervision" (2021), addresses a critical bottleneck in modern AI: the prohibitive cost of meta-training. Huang’s key contribution is a novel framework that enables policies to be meta-trained on static, offline datasets—labeled only once with rewards—dramatically reducing the need for expensive online interactions. By introducing online self-supervision, her method allows these pre-trained models to adapt to new tasks with orders of magnitude less data than standard RL, while still achieving robust performance. This work has garnered 8 citations and is foundational for researchers seeking sample-efficient, scalable learning systems. Huang’s research is pivotal for real-world applications where data collection is costly or risky, such as robotics and healthcare. Her achievements highlight a rare ability to bridge theoretical rigor with practical efficiency, making her a rising voice in the push toward truly autonomous, data-efficient agents.

Research Focus

Key Achievements

1
H-Index
1
Papers
8
Total Citations
8
Avg Citations/Paper
🏆 Most Cited Paper
Offline Meta-Reinforcement Learning with Online Self-Supervision
8 citations · 2021
📈 Most Prolific Year: 2021 (1 Papers)
🤝 Key Collaborators: 4
🏛 Institutions: University of California, Berkeley

Top Papers

  1. 1

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 12 days ago