Gyeonghoon Kim

Korea Advanced Institute of Science and Technology

Papers

Total Citations

H-Index

About

Gyeonghoon Kim is a leading figure in energy-efficient artificial intelligence hardware, with a focus on embedded neuro-fuzzy accelerators and mixed-mode processor design. His most cited work, "A 57mW embedded mixed-mode neuro-fuzzy accelerator for intelligent multi-core processor" (2011, 18 citations), pioneered low-power architectures that enable real-time AI functions—such as object detection, recognition, and human-computer interfaces—on portable devices like smartphones and robots. By integrating neural networks and fuzzy logic into compact, mixed-signal circuits, Kim demonstrated how to achieve high-performance AI processing within strict power budgets, a critical challenge for mobile and embedded systems. His contributions have helped bridge the gap between software-based AI algorithms and hardware implementation, influencing subsequent research in energy-efficient deep learning accelerators. Kim’s work is particularly notable for its practical focus on deploying intelligent capabilities in resource-constrained environments, making him a key innovator in the field of low-power AI hardware. His research continues to inspire advances in embedded machine learning and neuromorphic computing.