Chi-Chung Chen

Title: Innovations of Chi-Chung Chen

Introduction

Chi-Chung Chen is a prominent inventor based in Taipei, Taiwan. He has made significant contributions to the field of graphics processing and deep neural networks. With a total of nine patents to his name, Chen's work focuses on enhancing memory utilization and optimizing computational processes.

Latest Patents

One of his latest patents is titled "Method, system and storage media for training a graphics processing neural network with a patch-based approach." This invention involves a method for training a graphics processing neural network by calculating overlapping and invalid sizes of outputs from multiple feature extraction layers. The method divides input images into patches during both forward and backward passes, ensuring streamlined operations and improved memory utilization.

Another notable patent is "Enhancing memory utilization and throughput in executing a computational graph." This method enhances memory utilization and throughput of a computing platform during deep neural network training. It involves calculating memory sizes for operations in a computational graph and executing them in parallel groups, optimizing memory usage and significantly increasing throughput.

Career Highlights

Chi-Chung Chen has worked with notable companies such as Ritdisplay Corporation and Aetherai IP Holding LLC. His experience in these organizations has contributed to his expertise in the field of neural networks and graphics processing.

Collaborations

Some of his coworkers include Tien Rong Lu and Meng-Chieh Liao, who have collaborated with him on various projects.

Conclusion

Chi-Chung Chen's innovative patents and contributions to the field of graphics processing and deep learning demonstrate his commitment to advancing technology. His work continues to influence the development of efficient computational methods in neural networks.