BEIJING, China

Zhiming Gui

This inventor holds 1 USPTO granted patent. Top assignees: Beihang University, Beijing University of Technology. Active years: 2026.


Average Co-Inventor Count = 1.0

ph-index = 1


Company Filing History:


Years Active: 2026

Loading Chart...
1 patent (USPTO):Explore Patents

Title: Zhiming Gui: Innovator in Deep Learning Security

Introduction

Zhiming Gui is a prominent inventor based in Beijing, China. He has made significant contributions to the field of neural network security, particularly through his innovative patent in deep learning technology. His work focuses on enhancing the security of neural networks against potential backdoor attacks.

Latest Patents

Zhiming Gui holds a patent titled "Deep learning backdoor attack method and device based on ordinal network and medium." This invention discloses a method and device that utilizes an ordinal network to improve the concealability of poisoning samples in deep learning systems. The method involves obtaining a training sample image, generating an ordinal network to represent the image's structure, and creating a trigger using this network. This approach not only enhances the effectiveness of backdoor attacks but also contributes to the academic research on hidden backdoor attack defense methods. He has 1 patent to his name.

Career Highlights

Throughout his career, Zhiming Gui has worked at notable institutions such as Beihang University and Beijing University of Technology. His experience in these prestigious universities has allowed him to collaborate with other experts in the field and further his research in deep learning and neural network security.

Collaborations

Some of his notable coworkers include Yunfeng Lu and Chao Li. Their collaboration has likely contributed to the advancements in the research and development of deep learning technologies.

Conclusion

Zhiming Gui's innovative work in deep learning security showcases his commitment to advancing technology in a critical area. His contributions are paving the way for future research and development in neural network safety.

Profile summary based on public USPTO records.
Please report any incorrect information to [email protected]
Loading…