Palo Alto, CA, United States of America

Doe Hyun Yoon

This inventor holds 25 USPTO granted patents and 7 published patent applications, plus 1 CIPO patent and 9 EPO patents, primarily in Memory Architecture. Top assignees: Hewlett Packard Enterprise Development LP, Google Inc., Hewlett-Packard Development Company, L.p.. Active years: 2014-2025.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Doe Hyun Yoon. Retrieved from https://idiyas.com/inventor/doe-hyun-yoon-5rqlgl06

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

Calculating rank
USPTO Granted Patents = 25 

% Patents Active = 84.0


 

Average Co-Inventor Count = 3.3

ph-index = 4

Forward Citations = 45(Granted Patents)


Location History:

  • Austin, TX (US) (2014 - 2015)
  • San Jose, CA (US) (2015 - 2020)
  • Palo Alto, CA (US) (2017 - 2020)
  • Foster City, CA (US) (2022 - 2024)

Company Filing History:


Years Active: 2014-2025

Loading Chart...
Loading Chart...
Loading Chart...
Areas of Expertise:
Neural Network Accelerator
Systolic Array
Error Checking
Machine Learning Models
Memory Controllers
Smart Memory Buffers
Nonvolatile Memory
Local Checkpointing
Data Representation
Computer Power Management
Multi-Level Cache
Memory Error Identification
25 patents (USPTO):Explore Patents

Title: Doe Hyun Yoon: Innovator in Systolic Array Technology

Introduction

Doe Hyun Yoon is a prolific inventor based in Palo Alto, California, with an impressive portfolio of 24 patents that reflect his contributions to the field of technology. His innovative work focuses on enhancing the efficiency and accuracy of computational processes, particularly through the application of systolic arrays in deep neural networks.

Latest Patents

Among his latest patents, Yoon has developed two significant innovations: the "Multiplier and Adder in Systolic Array" and "Error Checking for Systolic Array Computation." The first patent describes systems and techniques for designing multiply-and-accumulate (MAC) units that can perform matrix multiplication more effectively by leveraging the inherent characteristics of systolic arrays. The method allows for the reuse of scalar values during the computation process, enhancing the overall performance of deep neural network accelerators. The second patent focuses on a computation unit that implements error detection mechanisms within the systolic array, allowing checksums to be computed and compared in real-time, thus ensuring accurate data processing without delaying operations.

Career Highlights

Doe Hyun Yoon has had an illustrious career, contributing his expertise to renowned companies such as Hewlett Packard Enterprise Development and Google Inc. His work in these organizations has significantly advanced the field of computational technology, particularly through his research and development of systems optimized for large-scale data processing tasks.

Collaborations

Throughout his career, Yoon has collaborated with esteemed colleagues, including Norman Paul Jouppi and Jichuan Chang. Their joint efforts have led to the development of groundbreaking techniques that push the boundaries of computational efficiency and accuracy, particularly in the realm of artificial intelligence and machine learning.

Conclusion

With a strong foundation in innovative technology and a commitment to advancing computational methodologies, Doe Hyun Yoon continues to make significant strides in the field. His patents not only demonstrate his technical prowess but also highlight the importance of collaboration and continuous improvement in the rapidly evolving landscape of technology.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…