Yen-Tien Wu

Title: Yen-Tien Wu and Optical Inspection-Based Defect Classification.

Introduction

Yen-Tien Wu is an inventor based in Castro Valley, California, whose patent work addresses automated defect classification in semiconductor manufacturing. Wu is associated with Applied Materials, Inc. The available information identifies 1 patent.

Latest Patents

Wu’s patent, “Optical Inspection-Based Automatic Defect Classification,” describes systems and techniques for efficiently inspecting semiconductor manufacturing samples. The disclosed approach collects optical inspection data from training samples containing multiple defects and uses that data to create a training set with defect descriptions, images, and ground-truth classifications.

The training set is used to train multiple machine-learning classifiers to predict defect classifications. The system then compares those predictions with the ground-truth classifications, selects classifiers that meet specified accuracy criteria, and applies the selected classifiers to classify defects in semiconductor manufacturing samples.

Career Highlights

Wu’s identified patent focuses on applying machine learning to optical inspection, with the aim of supporting accurate and efficient defect classification in semiconductor manufacturing. The available information associates Wu’s work with Applied Materials, Inc.

Collaborations

The listed coworkers are Navneet Kumar Singh and Arun Ramaswamy Srivatsa. The provided information does not specify their roles in the patent or the nature of their collaboration with Wu.

Conclusion

Yen-Tien Wu’s identified patent presents a machine-learning approach to classifying semiconductor defects using optical inspection data, trained classifiers, and accuracy-based selection.