ECE 423

Implementation of machine learning algorithms on resource-constrained hardware platforms, e.g., IoTs, AR/VR, autonomous vehicles, and biomedical devices. Focused on structure and function of deep neural networks (DNNs); fixed-point requirements for inference and training; lightweight DNNs; algorithm-to-architecture mapping to minimize energy and delay; digital and in-memory AI accelerator architectures. Coursework to include bi-weekly assignments with analysis, Python and Verilog programming and final project on design (undergraduates) or research (graduate students). Course Information: 3 undergraduate hours. 4 graduate hours. Prerequisite: ECE 313 and ECE 342.