Research
Our research develops mathematical models, optimization algorithms, and AI-driven decision-making frameworks for large-scale cyber-physical systems. We work at the intersection of artificial intelligence, operations research, optimization, and game theory to design intelligent, sustainable, resilient, and trustworthy systems that operate efficiently under uncertainty.
Our work combines optimization, machine learning, economics, and multi-agent decision-making to address fundamental challenges in AI infrastructure, cloud and edge computing, electric power systems, transportation, and digital marketplaces. A major focus of our recent research is the development of optimization methods for sustainable AI infrastructure, including large language model (LLM) training and inference, distributed GPU resource management, and the interaction between AI data centers and the electric power grid.
Current Research Directions
Sustainable AI infrastructure and LLM training/inference optimization
Grid-aware AI systems and AI–power grid interaction
Quantum optimization and quantum machine learning
Decision-making under uncertainty (stochastic, robust, distributionally robust, online, and decision-dependent optimization)
Multi-agent systems, resource allocation, and fairness-aware optimization
Market design, mechanism design, dynamic pricing, and network economics
Privacy-preserving distributed optimization and federated learning
Cloud, edge, and distributed AI systems
Electric vehicle charging infrastructure planning and operation