Fang Cao

Title: Innovations of Fang Cao in GPU Backend Training and Solar Cell Technology

Introduction

Fang Cao is a notable inventor based in Suzhou, China. He has made significant contributions to the fields of deep learning and solar cell technology. With a total of 2 patents, his work focuses on enhancing the performance of neural network training and improving solar cell efficiency.

Latest Patents

Fang Cao's latest patents include the "NGraph-based GPU backend distributed training method and system." This invention provides a method for performing neural network distributed training using an nGraph GPU backend. The process involves receiving a training request, obtaining training data, and utilizing a Nvidia® Collective multi-GPU Communication Library (NCCL) for efficient communication during training. This innovation addresses the urgent need for improved performance in deep learning network training.

Another significant patent is the "Texturization method of silicon wafers, product obtained therefrom and preparation method of solar cells." This method involves creating a nano-textured surface on silicon wafers, which enhances the efficiency of solar cells. The process includes superimposing two silicon wafers and performing nano-textured surface etching, which reduces fragmentation during the detaching process.

Career Highlights

Fang has worked with several companies, including Inspur Suzhou Intelligent Technology Co., Ltd and Csi Cells Co., Ltd. His experience in these organizations has contributed to his expertise in innovative technologies.

Collaborations

Fang has collaborated with notable coworkers, including Li Wang and Zhiyong Qiu. Their teamwork has likely fostered advancements in their respective fields.

Conclusion

Fang Cao's contributions to GPU backend training and solar cell technology demonstrate his innovative spirit and commitment to enhancing technological performance. His patents reflect a deep understanding of complex systems and a drive to improve efficiency in critical areas.