Greetings! 

I am Vipul, a fourth year PhD scholar at School of Industrial Engineering at Purdue University. I work in the field of learning-based decision-making algorithms (RL and Real-time Control) under hard constraints (e.g. safety) under the genius supervision of Sivaranjani Seetharaman. 

In RL, I am currently working on model-free policy gradient algorithms with convergence guarantees for intelligently parameterized RL policy under hard state-action coupled (e.g. safety) constraints. In Controls, I have been working on risk-tunable safe control design by learning environmental uncertainties and dynamical uncertainties. 

I am also working on RL for optimal transport under ODE constraints, multi-agent safe controls/RL and LLMs, as part of my long term research objectives of developing smart learning methods.

My broad areas of interest are:

• Safe Learning Algorithms (RL and Control)

• Optimality (PL condition) and Convergence (Model-free Policy Gradients) of Learning Algorithms

• Smart Policy Geometries (e.g. PID policies) for RL

• Safe Control under Uncertainty (e.g. Heavy-tailed)

• Robotics

During my 1st year at Purdue, I joined United States Army Research Laboratory for the summer of 2023 as an intern, and, worked on theory & applications in safe, and multi-agent reinforcement learning, with Wesley Suttle and Brian Sadler.

Before Purdue, I was an optimization/software engineer at Siemens India, working on mixed-integer progamming algorithms for electricity market operations. I also conducted research in the field of RL applied on small-grids, and resource allocation via DERs, leading to to an IP.

I completed my master's from Department of Electrical Engineering at Indian Institute of Science(IISc) where I had the privilege of starting my research under the guidance of Pavankumar Tallapragada, where I worked on observability under privacy concerns and briefly on multi-agent robotics.

I enjoy photography, swimming, singing, playing guitar, table-tennis and badminton. Following my inclinations towards the brain, I am also in my fifth year of meditating everyday, over 16,000 minutes now, and find it crucial exercise in the current age. Furthermore, I have been nurturing my aptitude towards cognitive sciences and I hope to apply mathematical tools in cognitive sciences, contributing towards theoretical frameworks for the human brains & behaviors.  Combined with my existing research, I believe that it is one of the best ways for me to contribute back to society.