Shu-Yin Ho

Title: Innovations of Shu-Yin Ho

Introduction

Shu-Yin Ho is a notable inventor based in Taichung, Taiwan. He has made significant contributions to the field of artificial neural networks, holding a total of 3 patents. His work focuses on enhancing computational methods and data representation in neural networks.

Latest Patents

One of his latest patents is an "Operation method for artificial neural network." This method involves dividing input information into multiple sub-inputs and expanding kernel information to generate expanded kernel information. It utilizes Fast Fourier Transform (FFT) to create frequency domain representations, which are then processed to extract features effectively. Another significant patent is the "Neural network computation method using adaptive data representation." This method is designed for processors to perform multiply-and-accumulate operations on a memory with a crossbar architecture. It adaptively divides input cycles into sub-cycles, optimizing the computation process.

Career Highlights

Shu-Yin Ho is currently employed at Macronix International Co., Ltd., where he continues to innovate in the field of neural networks. His work has been instrumental in advancing the capabilities of artificial intelligence and machine learning technologies.

Collaborations

He collaborates with talented coworkers such as Yuan-Hao Chang and Hsiang-Pang Li, contributing to a dynamic research environment that fosters innovation.

Conclusion

Shu-Yin Ho's contributions to artificial neural networks and his innovative patents highlight his role as a leading inventor in the field. His work continues to influence advancements in technology and artificial intelligence.