Dr. Tao Han, Assistant Professor of Electrical and Computer Engineering, won the prestigious National Science Foundation's (NSF) Faculty Early Career Development (CAREER) award. Dr. Han has been awarded $449,779 for his CAREER proposal AutoEdge: Deep Reinforcement Learning Methods and Systems for network Automation at Wireless Edge, for the award duration May 15, 2021 - April 30, 2026.
Dr. Han’s CAREER proposal is focused on applications of deep reinforcement learning (DRL) for achieving a diverse set of performance features in future wireless edge computing networks. His approach involves development of multiple logical networks known as network slices to meet a wide range of performance requirements that are typical of future heterogenous services planned for such networks. His project proposes to study the design of states, reward functions, training algorithms, and neural networks of domain-specific DRL, develop methods of handling various constraints in DRL-based end-to-end resource orchestration to avoid constraint violations, and design context-aware multi-agent DRL methods to leverage domain knowledge of wireless edge computing to improve the learning efficiency of DRL. The project also involves development of policy distillation methods to address the DRL deployment issues caused by the divergence between network simulations and real network systems, and design cross-scale knowledge transfer methods to address the mismatch of the dimensions of small-scale testbeds and large-scale wireless edge computing systems. Dr. Han plans to work with industrial partners such as Nokia Bell Labs and InterDigital Communications for investigating DRL for edge computing. Dr. Han’s plans include increasing the number of under-represented minority students in the graduate programs in ECE.