logo

For the Inventor, By the Inventor

Kiryat Motzkin, Israel

Ehud Cohen

This inventor holds 36 USPTO granted patents and 6 published patent applications and 2 EPO patents, primarily in Machine Learning Optimization. Top assignee: Intel Corporation. Active years: 2003-2026.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Ehud Cohen. Retrieved from https://idiyas.com/inventor/ehud-cohen-vpodhojm

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

USPTO Granted Patents = 36 

% Patents Active = 69.4

 

Average Co-Inventor Count = 5.6

ph-index = 7

Forward Citations = 240(Granted Patents)


Location History:

  • Kiryat, IL (2013)
  • Motskin, IL (2012 - 2014)
  • Kiryat Motskin, IL (2003 - 2023)
  • Kiryat Matskin, IL (2024)

Company Filing History:

goldMedal36 out of 56,365 
 
Intel Corporation
 patents

Years Active: 2003-2026

Loading Chart...
Loading Chart...
Areas of Expertise:
Data Fetch Errors
Matrix Compression
Memory Management
Convolutional Neural Networks
Dynamic Memory Storage
Power Consumption
Transaction Monitoring
Cache Coherency
Data Element Comparison
Graph Compression
Context Switching
GPU Implementation
36 patents (USPTO):Explore Patents
1. 12675681 - Smart memory handling and data management for machine learning networks (2026-07-07),
2. 12189479 - Apparatus and method for detecting and recovering from data fetch errors (2025-01-07),
3. 12131507 - Low rank matrix compression (2024-10-29),
4. 12131250 - Inner product convolutional neural network accelerator (2024-10-29),
5. 11886984 - Variable precision and mix type representation of multiple layers in a network (2024-01-30),
6. 11763140 - Smart memory handling and data management for machine learning networks (2023-09-19),
7. 11620766 - Low rank matrix compression (2023-04-04),
8. 11600035 - Sub-graph in frequency domain and dynamic selection of convolution implementation on a GPU (2023-03-07),
9. 11250610 - Sub-graph in frequency domain and dynamic selection of convolution implementation on a GPU (2022-02-15),
10. 11113053 - Data element comparison processors, methods, systems, and instructions (2021-09-07),
11. 11093822 - Variable precision and mix type representation of multiple layers in a network (2021-08-17),
12. 11087206 - Smart memory handling and data management for machine learning networks (2021-08-10),
13. 11048587 - Apparatus and method for detecting and recovering from data fetch errors (2021-06-29),
14. 11037330 - Low rank matrix compression (2021-06-15),
15. 10817567 - Techniques for information graph compression (2020-10-27),

Title: Ehud Cohen: Innovator and Patent Holder in Data Processing Technologies

Introduction

Ehud Cohen, a prolific inventor based in Kiryat Motskin, Israel, has made significant contributions to the field of data processing technologies. With an impressive portfolio of 35 patents, Cohen's innovations focus on improving data integrity and efficiency within processor architectures.

Latest Patents

Cohen's recent patents showcase his expertise in addressing critical issues in data processing. One of his notable inventions is an apparatus and method for detecting and recovering from data fetch errors. This invention involves an instruction processing apparatus that identifies errors during the data fetch operation and ensures that such errors do not compromise the processor's architectural state during the retirement stage.

Another significant patent is related to low rank matrix compression. This innovation incorporates hardware logic to execute a lossy compression algorithm, which is essential for enhancing data handling in convolutional neural networks (CNNs). By utilizing a data transform and quantization process, this technology optimizes data storage and processing efficiency in machine learning applications.

Career Highlights

Ehud Cohen is currently affiliated with Intel Corporation, a leading technology company known for its advancements in semiconductor manufacturing and computing technologies. His work at Intel has placed him at the forefront of innovation in data processing, where he continues to push boundaries through his inventions.

Collaborations

During his career, Cohen has collaborated with notable colleagues, including Uzi Sarel and Tomer Schwartz. These partnerships have been instrumental in fostering a collaborative environment that drives research and innovation within the organization, leading to the development of transformative technologies.

Conclusion

Ehud Cohen stands out as a remarkable inventor whose contributions are shaping the future of data processing technologies. With a robust patent portfolio and ongoing collaborations, he continues to drive innovation at Intel Corporation and within the broader technology landscape. His work not only addresses current challenges but also sets the stage for future advancements in data integrity and processing efficiency.

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…