Jun Yong Cheon

Title: Jun Yong Cheon and Neural Network Quantization

Introduction

Jun Yong Cheon is an inventor associated with Yonsei University in Seoul, South Korea. His patent work addresses quantization techniques for artificial neural networks and image processing.

Latest Patents

Cheon is associated with 1 patent: “Quantization Apparatus and Method for Artificial Neural Network and Image Processing Device Having the Same.” The invention divides an input matrix into channels, analyzes the distribution of values in each channel, and selects a quantizer for each one. It then quantizes the channel values using the selected quantizers, aiming to improve operational efficiency while limiting performance degradation caused by quantization errors.

Career Highlights

Cheon’s listed patent reflects work at the intersection of neural network computation, quantization, and image processing. His affiliation with Yonsei University connects this invention to the university’s research and innovation activities.

Collaborations

Bum Sub Ham and Jae Hyeon Moon are identified as Cheon’s coworkers. The available information does not specify their roles in the patent or the nature of their collaboration.

Conclusion

Jun Yong Cheon’s patent presents a channel-based approach to neural network quantization, with the goal of balancing computational efficiency and model performance.