Yangrui Li

Title: Innovations of Yangrui Li in Deep Learning and Code Classification

Introduction

Yangrui Li is a prominent inventor based in Wuhan, China. He has made significant contributions to the field of deep learning, particularly in the area of code classification. His innovative approach has led to the development of a unique method that enhances the interpretability of deep-learning models.

Latest Patents

Yangrui Li holds a patent for a "Sample-difference-based method and system for interpreting deep-learning model for code classification." This patent describes a method that includes off-line training of an interpreter by constructing code transformations for each code sample in a training set to generate difference samples. The process involves generating difference samples through feature deletion and code snippets extraction, calculating feature importance scores, and inputting these into a neural network to create a trained interpreter. The on-line interpretation of code samples utilizes this trained interpreter to extract important features and identify the most contributive training samples, ultimately generating interpretation results for the object samples.

Career Highlights

Yangrui Li is affiliated with Huazhong University of Science and Technology, where he continues to advance research in deep learning and its applications. His work has garnered attention for its innovative approach to interpreting complex models, making significant strides in the field.

Collaborations

Yangrui Li collaborates with notable colleagues, including Zhen Li and Ruqian Zhang. Their combined expertise contributes to the advancement of research and innovation in their respective fields.

Conclusion

Yangrui Li's contributions to deep learning and code classification exemplify the impact of innovative thinking in technology. His patent and ongoing research continue to shape the future of interpretability in machine learning models.