Johnny Rhe

I am currently pursuing a Ph.D. degree in the Department of Electrical and Computer Engineering at Sungkyunkwan University. My research focuses on algorithm-hardware co-design for efficeint In-Memory Computing (IMC)-based Deep/Convolutional Neural (D/CNNs) inference. I have worked on a wide range of topics including mapping optimization via pruning and compression techniques, reliability enhancement on analog IMC arrays, and a software simulation framework. Recently, I have expanded my research interests to transformer-based models like Vision Transformer (ViT) and Large Language Models (LLMs), aiming to optimize their performance and inference efficiency in IMC systems. To achieve this, I am exploring various hardware-aware optimization techniques, with the ultimate goal of enabling scalable and energy-efficeint deployment of these models on IMC systems. In the long term, my goal is to bridge algorithmic innovation with emerging hardware platforms, contributing to the realization of scalable, low-power AI systems for real-world applications.