Shuibing He

Title: Innovations by Shuibing He: A Neural Network Approach to Distributed Data Routing

Introduction

Shuibing He is an accomplished inventor based in Hangzhou, China. He has made significant contributions in the field of neural network computing, particularly through his innovative work that focuses on data routing. His research not only highlights his expertise but also addresses complexities in distributed design and development.

Latest Patents

Shuibing He holds a patent for a "Neural network computing-oriented modeling method and apparatus for distributed data routing." This patent presents a novel approach that includes several key steps: First, it involves designing the distributed attributes of a physical tensor by abstracting the mapping relationship between a logic tensor and a physical tensor. These attributes consist of three categories: a broadcast attribute, a scatter attribute, and a local reduction attribute. Next, the method deduces the distributed attribute of an output tensor based on the known distributed attributes of the input tensor. Finally, it judges whether an intermediate communication primitive needs to be inserted to achieve the desired attributes of a local physical tensor. This approach simplifies the difficulties often encountered in distributed design and promotes the development of applications for large deep neural network models.

Career Highlights

Shuibing He is currently associated with Zhejiang Lab, where he applies his expertise in neural network computing and data routing methodologies. His innovative patent underscores his commitment to enhancing computational processes by optimizing data management systems.

Collaborations

In his professional journey, Shuibing He collaborates with esteemed colleagues such as Hongsheng Wang and Hujun Bao. Their collective expertise fosters a dynamic research environment that aims to push the boundaries of neural network applications and their real-world implications.

Conclusion

Shuibing He's work exemplifies the impact of innovation in the realm of neural networks and distributed data routing. Through his patent, he demonstrates a keen understanding of the challenges in this field and offers practical solutions that contribute to the advancement of technology. His role at Zhejiang Lab and collaborations with notable peers position him as a pivotal figure in the ongoing evolution of neural network computing.