Arec Jamgochian

Papers

Total Citations

H-Index

About

Arec Jamgochian is a rising star in the field of artificial intelligence, whose work is pushing the boundaries of safe and reliable decision-making under uncertainty. His research centers on constrained planning in partially observable environments, a critical area for deploying autonomous systems in the real world. Jamgochian’s major contribution lies in developing novel algorithms for Constrained Partially Observable Markov Decision Processes (CPOMDPs), which generalize traditional planning by requiring agents to maximize rewards while strictly adhering to hard cost constraints—a necessity for safety-critical applications like autonomous driving or robotics. His most cited work, "Constrained Hierarchical Monte Carlo Belief-State Planning" (2024), tackles the immense computational challenge of online CPOMDP planning in large or continuous state spaces. By introducing a hierarchical, Monte Carlo-based approach, this paper offers a scalable solution to a problem previously considered intractable, earning early recognition with 2 citations. Though early in his career, Jamgochian’s contributions are already shaping the future of safe AI, providing a rigorous framework for agents that must act optimally without compromising safety.