Project With Liangliang Wang

Sequential Monte Carlo Methods for Model Based Reinforcement Learning

This project is to study model-based reinforcement learning for some toy examples using sequential Monte Carlo (SMC) methods. The goal is to investigate how to frame a reinforcement learning (RL) problem as an inference problem as a probabilistic graphical model and conduct statistical inference using SMC. Applicants for this project should be good at software development in popular programming languages (e.g. R, Python, Java, javascript) and Bayesian statistical inference with Markov chain Monte Carlo and/or SMC.