Po-Chen Lin

Title: Po-Chen Lin: Innovator in Neural Network Training

Introduction

Po-Chen Lin is a prominent inventor based in Hsinchu, Taiwan. He has made significant contributions to the field of neural networks, particularly through his innovative methods of training these complex systems. His work is recognized for its potential to enhance the efficiency and effectiveness of neural network applications.

Latest Patents

Po-Chen Lin holds a patent titled "Method of neural network training using floating-point signed digit representation." This patent describes a method for training a neural network that includes multiple neural network weights and neurons. The method utilizes floating-point signed digit numbers to represent each of the neural network weights. Specifically, the mantissa of each weight is represented by multiple mantissa signed digit groups, while the exponent is represented by an exponent digit group. This innovative approach allows for weight adjustment computation and neural network inference computation using the exponent digit group and at least one of the mantissa signed digit groups. He has 1 patent to his name.

Career Highlights

Po-Chen Lin is affiliated with National Taiwan University, where he continues to advance research in neural networks and artificial intelligence. His academic background and ongoing research contribute to the development of cutting-edge technologies in this rapidly evolving field.

Collaborations

One of his notable collaborators is Tzi-Dar Chiueh, with whom he has worked on various projects related to neural networks and their applications.

Conclusion

Po-Chen Lin's innovative methods in neural network training represent a significant advancement in the field of artificial intelligence. His contributions are paving the way for more efficient neural network applications, showcasing the importance of research and innovation in technology.