Latent Planning as Optimization

We worked on researching Sequential decision making and Long horizon planning problems. We leverage tools from probabilistic inference, reinforcement learning, and latent optimizations, trying to create data efficient algorithms that are able to plan during training and adapt to novel tasks during inference on the fly by re-planning with fast inference methods.

We recieve advising from Professor Ying Nian Wu from UCLA Statistics & Data Science Department (we previously recieved advising from Professor Sicun Gao from UCSD's Computer Science & Engineering Department and dived into this problem from an search & optimization perspective with biological inspirations).