ECE PhD Proposal Review: Hongjia Li

PhD Proposal Review: Automation Design and DNN Acceleration Algorithms: From Software Implementation to Hardware Physical Design

Hongjia Li

Location: Zoom Link

Abstract: Deep learning has been growing at a fast pace in recent years and has been expanded into many application fields, with a wide range from image recognition, object detection to medical applications. Meanwhile, edging devices such as mobile devices are rapidly becoming the central computer and carrier for deep learning tasks. However, real-time execution has been limited due to the computation/storage resource constraints on these devices.
In this proposal review, I will dive into some aspects of DNN acceleration methods, including model compression techniques and software implementation optimizations. The goal is to achieve an unprecedented, real-time performance of large-scale neural network inference on edging devices. Additionally, an efficient physical design automation design is introduced for Adiabatic Quantum-Flux-Parametron (AQFP) circuits, meeting the unique features and constraints.