I am a Ph.D. candidate in Computer Science at Vanderbilt University, working under the supervision of Prof. Abhishek Dubey. My research interests include decision-making under uncertainty, planning, machine learning, and reinforcement learning, with a focus on applications in autonomous cyber-physical systems. I am particularly interested in developing innovative approaches for decision-making in high-dimensional, stochastic, and non-stationary environments, as well as enhancing the safety and efficiency of autonomous vehicles.

Prior to my doctoral studies, I earned my M.S. in Computer Engineering from Northwestern University, where I worked with Prof. Qi Zhu on securing connected and autonomous vehicles. My master’s thesis on Sybil Attack Detection in VANET received the Northwestern Best MS Computer Engineering Thesis Award. I completed my B.S. in Computer Engineering with a dual degree in Computer Science from Rensselaer Polytechnic Institute. My research has been published in conferences such as ICLR, AAMAS and ICCPS, and I have been honored with awards including the Vanderbilt C.F. Chen Best Paper Runner-Up Award.

Research

My research focuses on developing intelligent decision-making systems for autonomous cyber-physical systems, with a particular emphasis on addressing challenges in complex, uncertain, and dynamic environments. The work spans several interconnected areas:

Decision Making with High Dimensional Action Space

• Latent Macro Action Planner: Developed for offline reinforcement learning, enabling efficient decision-making in high-dimensional, stochastic environments through learned temporally extended actions.

Decision Making in Non-Stationary Environments

• Adaptive Monte Carlo Tree Search: Enables safe exploration and online adaptation to changing dynamics in model-based reinforcement learning tasks.
• NS-Gym Toolkit: Provides standardized evaluation environments for online decision-making algorithms in dynamically changing settings.

Runtime Safety Assurance of Autonomous Vehicles

• Dynamic Simplex framework: Improves performance without compromising safety in autonomous systems through planning with multiple generative models in dynamic environments.
• Advanced sampling techniques: Enhances robustness of autonomous vehicle systems through high-risk scenario generation in AV testing.
• Automated testing framework: Evaluates autonomous vehicles under adversarial conditions in simulations.

Securing Connected and Autonomous Vehicles

• Hybrid GCN-RNN Model: Developed to detect Sybil attacks in connected vehicle networks.
• Dual Cyber-Physical Blockchain Framework: Created for efficient security in large-scale vehicular networks.

Find out publications for each of these projects.