๐ Paper accepted at the DynaFront Workshop, NeurIPS 2026 on risk-averse multi-population mean-field games.
PhD in Machine Learning at GeorgiaTech
Advised by Prof. Panagiotis Tsiotras
I study how intelligent agents learn to make decisions in complex, competitive environments. My research bridges reinforcement learning, game theory, and decision making โ with a focus on multi-agent systems at scale.
Studying strategic interactions between large numbers of rational agents using mean-field game frameworks and equilibrium concepts.
Developing algorithms for agents to learn optimal policies in complex, high-dimensional, partially observable environments.
Bridging data-driven methods with classical control systems, including differential flatness and Gaussian process approaches.
Recent happenings from my research journey.
๐ Paper accepted at the DynaFront Workshop, NeurIPS 2026 on risk-averse multi-population mean-field games.
๐ Paper accepted at NeurIPS 2026 (Main Conference Track): "Learning Large-Scale Competitive Team Behaviors with Mean-Field Interactions."
๐ Paper accepted at CDC 2026: "Robust Mean-Field Games with Risk Aversion and Bounded Rationality."
๐ค Presented at the ALA (Adaptive and Learning Agents) Workshop, AAMAS 2025 on competitive team behavior learning with mean-field games.
๐ Paper published at American Control Conference (ACC) 2024: "Data-Driven Control of Adaptive Cruise Control Systems using Differential Flatness and Gaussian Processes."
๐ Started PhD in Machine Learning at Georgia Institute of Technology under the supervision of Prof. Panagiotis Tsiotras.
We introduce a new paradigm: risk-averse multi-population mean-field games, where each population optimizes a worst-case expected reward over dynamically feasible ambiguity sets of mean-field flows of a subset of the other populations.
We study mean-field games under bounded rationality and risk-averse objectives, developing robust equilibria that account for uncertainty in initial distributions of the population.
We propose a scalable reinforcement learning approach for competitive team-games using mean-field approximations coupled with online opponent model estimation.
A data-driven control framework for ACC systems leveraging differential flatness to simplify trajectory planning and GPs for uncertainty quantification.
I am a Machine Learning PhD student at the Georgia Institute of Technology, Atlanta. I work in the Dynamics and Control Systems Laboratory (DCSL) under the guidance of Prof. Panagiotis Tsiotras. My research sits at the intersection of game theory, reinforcement learning, and decision making. I am particularly interested in how large populations of strategic agents can learn to cooperate or compete โ and how we can design systems that are safe, risk-averse and robust to uncertainties.
Before Georgia Tech, I earned my undergraduate and master's degrees at IIT Bombay where I worked on control theory and analysis, particularly, learning-based control and geometric mechanics.
When I am not working on mathemtical proofs or training neural networks (hyperparameter tuning yay!), you can find me running around Atlanta, lifting weights, scouting for coffee shops or cooking.
I'm always happy to discuss research ideas, collaborations, or just chat about AI and game theory!
DCSL โ Dynamics and Control Systems Laboratory
Atlanta, Georgia, USA