Minkyu Park

Hitting the Memory Wall

I am a research intern in the CSArch Lab at Korea University (advised by Prof. Gunjae Koo), preparing to begin my M.S. degree in Spring 2027. My research interests include computer architecture, memory systems, and energy-efficient system design.

Previously, I was a research intern at the PSAL Lab, POSTECH (advised by Prof. Gwangsun Kim), where I extended PyTorchSim with new MLIR-based operator support. Before that, I worked on Edge AI power modeling at the AIAC Lab, Hallym University (advised by Prof. Jeong-Gun Lee).

selected publications

Predicting Delay, Power, and Energy of Edge AI Inference with Ridge Regression

Minkyu Park, Jaemin Jeong, Jiho Cho, and Jeong-Gun Lee

In IEEE International Conference on Consumer Electronics (ICCE), 2026

Abs Bib

This work proposes a ridge-regression-based framework for predicting delay, power, and energy consumption of deep learning inference on edge devices, enabling rapid design-space exploration for consumer electronics such as smart cameras, wearables, XR devices, and healthcare systems.
@inproceedings{park2026predicting, title = {Predicting Delay, Power, and Energy of Edge AI Inference with Ridge Regression}, author = {Park, Minkyu and Jeong, Jaemin and Cho, Jiho and Lee, Jeong-Gun}, booktitle = {IEEE International Conference on Consumer Electronics (ICCE)}, year = {2026}, }

Performance and Energy Consumption Analysis of Deep Learning Models on Edge GPU Systems

Minkyu Park, Hongkwon Kim, Jaemin Jeong, and Jeong-Gun Lee

In Proceedings of the Annual Conference of KIPS, 2025

Abs Bib

@inproceedings{park2025edge,
  title = {Performance and Energy Consumption Analysis of Deep Learning Models on Edge GPU Systems},
  author = {Park, Minkyu and Kim, Hongkwon and Jeong, Jaemin and Lee, Jeong-Gun},
  booktitle = {Proceedings of the Annual Conference of KIPS},
  volume = {32},
  number = {1},
  year = {2025},
}