Pavan Kantharaju

My current research focus lies at the intersection of machine learning, computer games, AI planning, and AI plan and goal recognition. I am particularly interested in studying how machine learning can be used to construct plan knowledge models from training data, and how these models can subsequently be utilized by intelligent agents for reasoning about/over actions in a domain. My dissertation focused on learning models for hierarchical planning and plan recognition which capture how complex tasks decompose into executable plans in a variety of application domains, including open world computer games and Real-Time Strategy games. Outside my dissertation, I have also done research in knowledge tracing for a parallel programming educational game, and embedding learning using deep neural networks.

Other research interests: Outside of my main research interests, I also have a lot of interest in game playing and deep learning.

Past research: During my undergraduate studies, I worked on research related to computer security and networking.