This inventor holds 1 USPTO granted patent. Top assignee: Intel Corporation. Active years: 2020.
Company Filing History:
Years Active: 2020
Title: Innovations of Urs Koster in Neural Network Computation
Introduction
Urs Koster is an accomplished inventor based in San Diego, CA. He has made significant contributions to the field of neural network computation through his innovative patent. His work focuses on enhancing the efficiency of numerical representation in distributed matrix processor architectures.
Latest Patents
Urs Koster holds a patent titled "Dynamic management of numerical representation in a distributed matrix processor architecture." This patent describes a system that receives and executes a sequence of tensor instructions, which are essential for performing neural network computations. The system is designed as a multiprocessor architecture, enabling efficient hardware execution of these computations. A tensor instruction specifies a tensor computation that receives one or more input tensors to determine an output tensor. The system also manages the decimal position associated with multiple values of a tensor, performing computations to derive a plurality of values for the output tensor. Additionally, it collects statistics on these values to optimize the decimal position based on the gathered data. Koster's innovative approach has the potential to significantly improve the performance of neural network applications.
Career Highlights
Urs Koster is currently employed at Intel Corporation, where he continues to push the boundaries of technology. His work at Intel has allowed him to collaborate with some of the brightest minds in the industry, contributing to advancements in computational technologies.
Collaborations
Some of his notable coworkers include William Howard Constable and Luke James Hornof. Their collaborative efforts have further enriched the research and development environment at Intel Corporation.
Conclusion
Urs Koster's contributions to neural network computation through his patent demonstrate his innovative spirit and technical expertise. His work not only enhances computational efficiency but also paves the way for future advancements in the field.