Hongbin Yang

Title: Innovations of Hongbin Yang

Introduction

Hongbin Yang is a prominent inventor based in Jiangsu, China. He has made significant contributions to the field of neural networks and image recognition technologies. With a total of 3 patents to his name, Yang continues to push the boundaries of innovation in his industry.

Latest Patents

Yang's latest patents include a neural network compression method, apparatus and device, and storage medium. This innovative method allows for forward inference on target data using a target parameter sharing network, which results in an output feature map from the last convolutional module. The method extracts channel-related features from this output and inputs them into a target meta-generative network to predict an optimal network architecture, ultimately leading to a compressed neural network model. Another notable patent is an image recognition method and apparatus based on a systolic array. This method converts image feature information into a one-dimensional feature vector and allocates corresponding weight groups to each node in a trained three-dimensional systolic array model. The process involves performing multiply-accumulate operations in parallel to determine the article category contained in an image.

Career Highlights

Hongbin Yang is currently employed at Inspur Suzhou Intelligent Technology Co., Ltd. His work focuses on advancing technologies that enhance the efficiency and accuracy of neural networks and image recognition systems. His innovative approaches have garnered attention in the tech community.

Collaborations

Yang collaborates with talented individuals such as Gang Dong and Yaqian Zhao. Their combined expertise contributes to the development of cutting-edge technologies in their field.

Conclusion

Hongbin Yang's contributions to neural networks and image recognition are noteworthy. His innovative patents and collaborative efforts position him as a key figure in the advancement of technology.