Yung-Chen Tang

Title: Yung-Chen Tang: Innovator in Neural Network Calibration

Introduction

Yung-Chen Tang is a notable inventor based in Hsinchu, Taiwan. He has made significant contributions to the field of neural networks, particularly in the area of calibration techniques. His innovative work has led to the development of a patent that addresses key challenges in this domain.

Latest Patents

Yung-Chen Tang holds a patent titled "Joint input perturbation and temperature scaling for neural network calibration." This patent focuses on systems, computer-readable media, and computer-implemented methods for calibrating neural networks. The invention involves determining a universal perturbation value and a temperature scaling parameter based on a training data set. It also describes processing a testing data set using a neural network by applying the universal perturbation value and the temperature scaling parameter to the logits determined by the neural network. This patent represents a significant advancement in the calibration of neural networks, with potential applications in various fields.

Career Highlights

Throughout his career, Yung-Chen Tang has worked with prestigious organizations, including IBM and Tsinghua University. His experience in these institutions has allowed him to collaborate with leading experts and contribute to groundbreaking research in artificial intelligence and machine learning.

Collaborations

Yung-Chen Tang has collaborated with I-Hsin Chung, a fellow researcher in the field. Their partnership has fostered innovative ideas and advancements in neural network technologies.

Conclusion

Yung-Chen Tang's contributions to neural network calibration through his patent and collaborations highlight his role as an influential inventor in the field. His work continues to inspire advancements in artificial intelligence and machine learning technologies.