Education
- BE (Computer) & MSc (Computer Science), Tribhuvan University
Bio
Pravesh is a PhD student and Graduate Research Assistant at VAMPIR. His research lies at the intersection of computational game theory, optimization, and control. He is particularly interested in multilevel and hierarchical games such as Stackelberg games and integer programming games (IPGs), where decision-making becomes especially challenging due to nested objectives and combinatorial strategy spaces.
One of Pravesh’s key contributions has been in designing Monte Carlo–based methods for approximating local equilibria in multilevel games. Traditional concepts like Nash equilibrium often fail to scale in these contexts, so we have explored alternative solution approaches that preserve theoretical rigor while remaining computationally feasible. This line of work has been successfully applied to toll-setting problems, pursuit–evasion dynamics, and hierarchical optimization tasks, showing both scalability and robustness in real-world-inspired settings.
More recently, Pravesh co-authored research on Locally Optimal Integer Solutions (LOIS), a new framework for analyzing and solving integer programming games. LOIS offers a practical alternative to pure Nash equilibria, especially in large-scale problems where exhaustive equilibrium computation is not possible. With this framework, we are able to characterize equilibrium-like behavior in integer strategy spaces, enabling analysis of competitive systems in areas such as cybersecurity, transportation, and multi-agent planning.
Originally from Nepal, Pravesh brings with him a background that motivates his interest in solving problems with real-world societal impact. His long-term goal is to advance both the theory and application of game-theoretic optimization, contributing tools that not only deepen mathematical understanding but also directly support engineering and decision-making in complex, adversarial environments.
