Thomas Unterthiner

Title: Thomas Unterthiner: Innovator in Neural Network Technologies

Introduction

Thomas Unterthiner is a prominent inventor based in Berlin, Germany. He has made significant contributions to the field of computer vision through his innovative patents. With a total of 3 patents, Unterthiner is recognized for his work in developing advanced neural network technologies.

Latest Patents

Unterthiner's latest patents include groundbreaking methods and systems for processing images using neural networks. One of his notable inventions is a multi-layer perceptron-based computer vision neural network. This patent outlines methods, systems, and apparatus for processing images using mixer neural networks. The process involves obtaining images, determining image patches, and generating input sequences for neural network processing. Another significant patent focuses on processing images using self-attention based neural networks. This invention similarly details methods for obtaining images, determining image patches, and processing input sequences to characterize the images.

Career Highlights

Thomas Unterthiner is currently employed at Google Inc., where he continues to push the boundaries of technology in the field of artificial intelligence and computer vision. His work has been instrumental in advancing the capabilities of neural networks, making them more efficient and effective in image processing tasks.

Collaborations

Unterthiner has collaborated with notable colleagues in the field, including Neil Matthew Tinmouth Houlsby and Jakob D Uszkoreit. These collaborations have further enriched his research and development efforts, leading to innovative solutions in neural network applications.

Conclusion

Thomas Unterthiner is a key figure in the realm of neural network technologies, with a focus on enhancing image processing capabilities. His contributions through patents and collaborations continue to shape the future of computer vision.