George Petre

This inventor holds 25 USPTO granted patents and 3 published patent applications and 14 EPO patents, primarily in Neural Network Processing. Top assignee: Microsoft Technology Licensing, LLC. Active years: 2020-2024.

IDiyas Innovation Intelligence. (2026). Inventor Profile: George Petre. Retrieved from https://idiyas.com/inventor/george-petre

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

USPTO Granted Patents = 25 (Patent Applications = 3)

% Patents Active = 100.0

EPO Patents = 14 (Patent Applications = 18)

Average Co-Inventor Count = 6.2

ph-index = 3

Forward Citations = 23(Granted Patents)

Company Filing History:

25 out of 56,279

Microsoft Technology Licensing, LLC

patents

Years Active: 2020-2024

Loading Chart...

Areas of Expertise:

Neural Processing Element

Memory Bandwidth

Data Management

Power Efficiency

Dynamic Sequencing

Throughput Vector Dequantization

Artificial Intelligence

DNN Module

Workload Management

Memory Utilization

Virtualized Data Iterator

Performance Enhancement

25 patents (USPTO):Sort by Forward CitationsExplore Patents

1. 12154027 - Increased precision neural processing element (2024-11-26),

2. 11909422 - Neural network processor using compression and decompression of activation data to reduce memory bandwidth utilization (2024-02-20),

3. 11750212 - Flexible hardware for high throughput vector dequantization with dynamic vector length and codebook size (2023-09-05),

4. 11722147 - Dynamic sequencing of data partitions for optimizing memory utilization and performance of neural networks (2023-08-08),

5. 11604972 - Increased precision neural processing element (2023-03-14),

6. 11528033 - Neural network processor using compression and decompression of activation data to reduce memory bandwidth utilization (2022-12-13),

7. 11507349 - Neural processing element with single instruction multiple data (SIMD) compute lanes (2022-11-22),

8. 11494237 - Managing workloads of a deep neural network processor (2022-11-08),

9. 11487342 - Reducing power consumption in a neural network environment using data management (2022-11-01),

10. 11476869 - Dynamically partitioning workload in a deep neural network module to reduce power consumption (2022-10-18),

11. 11405051 - Enhancing processing performance of artificial intelligence/machine hardware by data sharing and distribution as well as reuse of data in neuron buffer/line buffer (2022-08-02),

12. 11341399 - Reducing power consumption in a neural network processor by skipping processing operations (2022-05-24),

13. 11256976 - Dynamic sequencing of data partitions for optimizing memory utilization and performance of neural networks (2022-02-22),

14. 11205118 - Power-efficient deep neural network module configured for parallel kernel and parallel input processing (2021-12-21),

15. 11182667 - Minimizing memory reads and increasing performance by leveraging aligned blob data in a processing unit of a neural network environment (2021-11-23),

As an AI assistant specialized in innovations, inventions, inventors, patent attorneys, assignees, and patents, I can provide you with an article on the inventor George Petre based on the data provided.

Title: Innovator Spotlight: George Petre - Revolutionizing Neural Networks in Redmond, WA

Introduction:

George Petre, a prolific inventor based in Redmond, WA, is making waves in the tech industry with his groundbreaking work in neural network processors. With an impressive portfolio of 24 patents to his name, Petre's contributions are reshaping the landscape of deep learning and artificial intelligence.

Latest Patents:

Petre's latest patents showcase his innovative approach to optimizing memory bandwidth utilization in neural network processors. His creation of a neural network processor using compression and decompression of activation data is revolutionizing the way data is processed, offering a more efficient and streamlined solution for memory management. Additionally, his work on flexible hardware for high throughput vector dequantization is enhancing the performance of neural networks, paving the way for faster and more effective data processing.

Career Highlights:

Currently working at Microsoft Technology Licensing, LLC, Petre has established himself as a key player in the tech giant's research and development division. With a keen eye for innovation and a passion for pushing the boundaries of technology, Petre continues to spearhead projects that have a lasting impact on the industry.

Collaborations:

Petre's success is not achieved in isolation. Collaborating with talented individuals such as Chad Balling McBride and Amol Ashok Ambardekar, Petre has been able to leverage his expertise and skills to bring cutting-edge solutions to fruition. Together, they form a dynamic team that is driving innovation forward and shaping the future of neural networks.

Conclusion:

In conclusion, George Petre's remarkable work in neural network processors is setting a new standard for efficiency and performance in the tech industry. With a passion for innovation and a dedication to excellence, Petre continues to push the boundaries of what is possible, leaving a lasting impact on the world of artificial intelligence.

Profile summary based on public USPTO records.