This inventor holds 2 USPTO granted patents and 7 published patent applications. Top assignee: Micron Technology Incorporated. Active years: 2026.
Company Filing History:
Years Active: 2026
Title: Innovations by Dustin Werran in Deep Learning Acceleration
Introduction
Dustin Werran is an accomplished inventor based in New York, NY (US). He has made significant contributions to the field of deep learning, particularly in the area of acceleration with mixed precision. With a total of 2 patents to his name, Werran is recognized for his innovative approaches to enhancing computational efficiency.
Latest Patents
Werran's latest patents focus on devices designed for deep learning acceleration with mixed precision. One of his inventions includes a device that features a first data port for receiving a map data segment and a second data port for a kernel data segment. This device is equipped with a precision mode port that indicates the input precision mode, which determines the word length for both the map and kernel data segments. Additionally, it incorporates a multiplier component that generates an output based on the input precision mode by multiplying the map and kernel data segments. Another patent describes a device with matrix-vector components that utilize vector-vector components to generate outputs based on various precision modes. This device is designed to optimize the accumulation of products through efficient calculations.
Career Highlights
Dustin Werran is currently employed at Micron Technology Incorporated, where he continues to push the boundaries of innovation in deep learning technologies. His work is instrumental in developing advanced solutions that enhance the performance of machine learning applications.
Collaborations
Werran collaborates with talented individuals such as Aliasger Tayeb Zaidy and Sen Ma, contributing to a dynamic team focused on groundbreaking advancements in technology.
Conclusion
Dustin Werran's contributions to deep learning acceleration with mixed precision exemplify his commitment to innovation. His patents reflect a deep understanding of computational efficiency and a drive to improve machine learning technologies.