- Ph.D. Student in Machine Learning and Control at Hybrid Robotics Lab, BAIR, UC Berkeley
- Working on Reinforcement Learning and Stochastic Control
Safe and Robust RL, Scalable RL, Stochastic Control, Optimization
- What is intelligence?
- How can agents achieve safety, robustness, stability, efficiency, and continual learning?
- Can agents performance a.s. monotone increase using any data stream? (= Is RL scalable?)
Self-Supervised RL, Offline RL, Off2On RL, Off-Policy Q-Learning, Stochastic Control with Trajectory Regularization, World Models, Deep Learning for Dynamical Systems