Neuromorphic Computing

This course offers an interdisciplinary perspective on neuromorphic computing across the full stack of computing. It examines fundamentals and learning of artificial neural networks (ANNs), discusses operational principles and learning in spiking neural networks (SNNs), and reviews their implementations in hardware. It presents several state-of-the-art learning algorithms such as converting ANN to SNN, spike timing dependent plasticity, evolutionary approaches, and reservoir computing. Hardware-aware neural architecture search and Bayesian optimization approaches are also covered to co-optimize algorithm-hardware in this full-stack computing framework. This course involves projects focusing on applications of neuromorphic computing in computational neuroscience, control and robotics, smart healthcare, and smart city design. Offered by Electrical & Comp. Engineering. May not be repeated for credit.