Su Gil Lee

Title: Su Gil Lee: Innovator in ReRAM-Based Deep Learning Accelerators

Introduction

Su Gil Lee is a notable inventor based in Ulsan, South Korea. He has made significant contributions to the field of deep learning, particularly through his innovative work on resistive random access memory (ReRAM) technology. His research focuses on enhancing the performance and reliability of deep learning accelerators.

Latest Patents

Su Gil Lee holds a patent for a "Stuck-at fault mitigation method for ReRAM-based deep learning accelerators." This patent outlines a method that addresses the challenges posed by stuck-at faults (SAF) in ReRAM hardware. The process involves confirming distorted output values using a correction data set from a pre-trained deep learning network. It also includes updating the average and standard deviation of a batch normalization layer and folding this layer into convolution or fully-connected layers to derive normal output values.

Career Highlights

Throughout his career, Su Gil Lee has worked with prestigious institutions such as King Abdullah University of Science and Technology and UNIST-Academy Industry Research Corporation. His work has been instrumental in advancing the capabilities of deep learning technologies.

Collaborations

Su Gil Lee has collaborated with notable colleagues, including Ahmed Mohamed Eltawil and Jong Eun Lee. These partnerships have contributed to the development of innovative solutions in the field of deep learning.

Conclusion

Su Gil Lee's contributions to ReRAM-based deep learning accelerators highlight his role as a leading inventor in the technology sector. His innovative methods for fault mitigation are paving the way for more reliable and efficient deep learning applications.