Xiaoyu Zhang

Title: Xiaoyu Zhang and Efficient Analysis of Machine-Learning Models.

Introduction

Xiaoyu Zhang is an inventor based in Davis, California, whose patent work addresses the analysis and comparison of machine-learning models. The supplied information identifies her as a woman and lists Robert Bosch as her company.

Latest Patents

Zhang is associated with 1 patent, titled “System and Method for Efficient Analyzing and Comparing Slice-Based Machine Learning Models.” The invention describes a computer-implemented method for evaluating a machine-learning model across data slices that share common attributes. It uses a shallow regressor to predict residuals, compares modified accuracy with the model’s original accuracy, and determines each slice’s effect before presenting the results through a graphical interface.

Career Highlights

The patent focuses on making it easier to identify how a machine-learning model performs on different subsets of input data. By estimating residuals and measuring changes in accuracy, the method can help users examine model behavior across selected slices.

Collaborations

The supplied information names Jorge Henrique Piazentin Ono and Liang Gou as Zhang’s coworkers.

Conclusion

Xiaoyu Zhang’s patent describes a structured approach to comparing machine-learning performance across data slices and communicating the resulting effects.