Biyi Fang

Title: Innovations of Biyi Fang

Introduction

Biyi Fang is an accomplished inventor based in Lansing, MI (US). He has made significant contributions to the field of deep learning and neural networks, holding a total of 9 patents. His work focuses on creating systems that enhance the efficiency and adaptability of neural networks, particularly in resource-constrained environments.

Latest Patents

One of Biyi Fang's latest patents is titled "Systems and methods for implementing flexible, input-adaptive deep learning neural networks." This patent describes innovative systems and methods for creating input-adaptive neural networks that adjust their complexity based on the input data, such as video streams. The approach utilizes fine-grained design and an automatic method for generating optimal network architectures through early-exit mechanisms. These modified neural networks can be implemented in various devices, including mobile vision systems like smartphones, drones, and head-mounted augmented reality devices.

Another notable patent is "Techniques for enabling on-device shape recognition." This patent outlines a data processing system that trains a convolutional neural network (CNN) to classify shapes from digital ink inputs. The first CNN is designed to process ink stroke information and determine if it represents a hand-drawn geometric shape. The system also compresses this CNN to create a second CNN that can operate on resource-constrained devices, allowing for efficient processing of user-captured ink stroke information.

Career Highlights

Biyi Fang has worked with prominent organizations, including Microsoft Technology Licensing, LLC and Michigan State University. His experience in these institutions has allowed him to develop and refine his innovative ideas in deep learning and neural networks.

Collaborations

Biyi Fang has collaborated with notable colleagues, including Mi Zhang and Sheng Yi. These partnerships have contributed to the advancement of his research and the successful development of his patented technologies.

Conclusion

Biyi Fang's contributions to the field of deep learning and neural networks are noteworthy. His innovative patents demonstrate his commitment to enhancing technology in practical applications. His work continues to influence the development of efficient and adaptable systems in various devices.