Xinpeng Feng

Title: **Xinpeng Feng: Pioneering Innovations in Neural Networks**

Introduction

Xinpeng Feng is a renowned inventor based in Shanghai, China, holding an impressive total of 26 patents. His work primarily focuses on advancements in technology, particularly related to neural networks and data processing.

Latest Patents

One of his latest patents, titled "Memory pre-allocation for forward calculation in a neural network," introduces a method for optimizing memory usage in neural network computations. The patent outlines a systematic approach to pre-allocating memory blocks for outputs from each layer of the network. This method involves obtaining a memory block identifier from a list of idle memory blocks, storing it in a mapping table, and handling new memory block identifiers efficiently. This innovation is pivotal in enhancing the efficiency of data processing within neural networks.

Career Highlights

Xinpeng Feng has made significant contributions to the tech industry through his roles at various companies. Notably, he has worked at Nextvpu (Shanghai) Co., Ltd. and Semiconductor Manufacturing International (Shanghai) Corporation, where he played a key role in developing cutting-edge technologies that push the boundaries of electronic devices.

Collaborations

Throughout his career, Xinpeng Feng has collaborated with talented individuals in the field. Among his notable coworkers are Ji Zhou and Haijiao Cai, who have worked alongside him on projects that have led to significant technological advancements.

Conclusion

Xinpeng Feng's contributions to the field of neural networks and data processing are noteworthy, as reflected in his extensive patent portfolio. His innovative methods continue to influence the way technology evolves, paving the way for future advancements in the industry.