Aditya Mahajan

Aditya Mahajan is a Professor in the Department of Electrical and Computer Engineering at McGill University, Montreal, Canada. He holds affiliations with prominent research institutes including Mila (Quebec Artificial Intelligence Institute), CIM (Center for Intelligent Machines), GERAD (Group for Research in Decision Analysis), and ILLS (International Laboratory for Low Temperature Science).

His research focuses on stochastic control, reinforcement learning, decentralized systems, and multi-agent decision-making. Key themes include analysis of Markov decision processes (MDPs), robust control strategies, and optimization in partially observable environments. He explores theoretical foundations of actor-critic algorithms, human-automation collaboration, and networked control systems with applications to robotics and autonomous systems.

Recent work emphasizes scalable methods for complex systems, including approximation techniques for MDPs with unbounded costs, delay-resistant stochastic approximation, and policy design in POMDPs. His contributions span both algorithmic innovations and rigorous theoretical analysis, bridging gaps between control theory and modern machine learning paradigms.

Major research directions include:

• Reinforcement learning in decentralized and partially observable settings
• Optimal control for human-machine teams
• Robustness and convergence properties of stochastic algorithms
• Multi-agent systems and mean-field games

His articles highlight advancements in control theory, with recent attention to:

• Decision-making under cognitive load
• Convergence guarantees for Q-learning variants
• Structural results for restless bandits and resource allocation

He participates in interdisciplinary collaborations through his affiliations, contributing to AI applications in robotics, energy systems, and networked control architectures.