Xinmin Wu

Title: Innovations by Inventor Xinmin Wu

Introduction

Xinmin Wu is an accomplished inventor based in Cary, NC, USA, known for his significant contributions to the field of computing systems. With a total of 16 patents to his name, Wu has been at the forefront of innovative technologies that enhance machine learning and classification models.

Latest Patents

Wu's recent patents demonstrate his expertise in developing sophisticated computing solutions. One of his notable inventions is the "Distributed Gaussian Process Classification Computing System." This advanced system efficiently trains a classification model using distributed training data. By utilizing worker indices to uniquely identify segments of a lower triangular matrix, the system computes kernel matrix blocks and objective function values, ultimately outputting model parameters for trained classification models.

Another significant patent focuses on "Cutoff Value Optimization for Bias Mitigating Machine Learning Training System with Multi-Class Target." This computing device is designed to train fair prediction models by defining optimal event cutoff values. The process includes repeated calculations involving fairness constraints, leading to the output of an optimal event cutoff value and the corresponding prediction model.

Career Highlights

Xinmin Wu has built a successful career at SAS Institute Inc., where he has played a crucial role in advancing machine learning technologies. His impressive patent portfolio reflects his deep understanding of computational methods and their applications in predictive modeling.

Collaborations

Throughout his career, Wu has collaborated with talented individuals such as Xiangqian Hu and Yingjian Wang. Their teamwork has likely contributed to the innovation and effectiveness of the projects they have tackled together, enhancing the capabilities of machine learning systems developed at SAS Institute Inc.

Conclusion

With 16 patents and a focus on innovative computing systems, Xinmin Wu continues to push the boundaries of technology in machine learning and classification. His contributions not only showcase his talent as an inventor but also signify a commitment to creating fair and efficient predictive models that can benefit various industries.