Ranveer Singh: “Combining Planning and Reinforcement Learning for Solving Relational Multiagent Domains”

My research has been on sequential decision making and temporal modelling. My most recent work has been on using planner as a centralized controller in conjunction with reinforcement learning systems to solve problems in relational multiagent domains. You can find more details about the work here.

Additionally, I am working on modelling the risk of Severe Neurological Injury for patient on ECMO using high frequency physiological and laboratory data. The problem is challenging due to high dimensionality and small sample size of the data. I have been looking at the use of Large Language Models as approximate data sources to improve the sample efficiency of such models.