About
Christopher Amato is Associate Professor in the Khoury College of Computer Sciences at Northeastern University. His research develops principled methods for multi-agent systems operating under uncertainty with limited communication, with applications in multi-robot coordination, autonomous systems, and artificial intelligence. He directs research on reinforcement learning approaches for decentralized control in complex environments.
His work integrates techniques from reinforcement learning, game theory, and probabilistic reasoning to enable efficient coordination in applications including disaster response, surveillance, and networked systems. Recent projects address cooperative multi-agent reinforcement learning (MARL) under partial observability, asynchronous learning frameworks, and robust multi-robot coordination. He co-organized the COMARL symposium on challenges in multi-agent reinforcement learning.
Dr. Amato earned his PhD from UMass Amherst under Shlomo Zilberstein, with postdoctoral research at MIT working with Leslie Kaelbling and Jonathan How. His publications include foundational work on decentralized partially observable Markov decision processes (Dec-POMDPs) and multi-agent reinforcement learning. He currently advises seven PhD students working on MARL theory and applications.
Research Areas:
- Partially observable reinforcement learning
- Multi-agent/robot systems
- Decision-making under uncertainty
- Scalable coordination algorithms
