logo

For the Inventor, By the Inventor

Tempe, AZ, United States of America

Jae-sun Seo

This inventor holds 33 USPTO granted patents and 10 published patent applications, primarily in In-Memory Computing. Top assignees: Arizona State University, International Business Machines Corporation, Columbia University. Active years: 2011-2026.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Jae-sun Seo. Retrieved from https://idiyas.com/inventor/jae-sun-seo

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

USPTO Granted Patents = 33 

% Patents Active = 66.7

Average Co-Inventor Count = 3.7

ph-index = 6

Forward Citations = 167(Granted Patents)


Location History:

  • Ann Arbor, MI (US) (2011)
  • White Plains, NY (US) (2012 - 2022)
  • Tempe, AZ (US) (2016 - 2023)

Company Filing History:

goldMedal14 out of 2,164 
 
Arizona State University
 patents
silverMedal13 out of 166,351 
 
International Business Machines Corporation
 patents
bronzeMedal5 out of 2,554 
 
Columbia University
 patents
4. 3 out of 56,322 
 
Intel Corporation
 patents
5. 1 out of 834,513 
Other
 patents
where one patent can have more than one assignee

Years Active: 2011-2026

Loading Chart...
Areas of Expertise:
In-Memory Computing
Deep Neural Networks
Biometric Security
Cascaded Computing
Static RAM
Resistive Memory
Neurosynaptic Module
Voltage Conversion
Neuromorphic Hardware
General-Purpose Circuits
Memory Compression
Signal Processing
33 patents (USPTO):Explore Patents
1. 12755152 - Method and system for a temperature-resilient neural network training model (2026-10-06),
2. 12488228 - Programmable in-memory computing accelerator for low-precision deep neural network inference (2025-12-02),
3. 11783875 - Circuits and methods for in-memory computing (2023-10-10),
4. 11775831 - Cascaded computing for convolutional neural networks (2023-10-03),
5. 11727261 - Static random-access memory for deep neural networks (2023-08-15),
6. 11698952 - Smart hardware security engine using biometric features and hardware-specific features (2023-07-11),
7. 11556779 - Cascaded computing for convolutional neural networks (2023-01-17),
8. 11501829 - Resistive random-access memory for embedded computation (2022-11-15),
9. 11355167 - Circuits and methods for in-memory computing (2022-06-07),
10. 11295201 - Time-division multiplexed neurosynaptic module with implicit memory addressing for implementing a neural network (2022-04-05),
11. 11170292 - Static random-access memory for deep neural networks (2021-11-09),
12. 11074498 - Static and dynamic precision adaptation for hardware learning and classification (2021-07-27),
13. 10827952 - Electrocardiographic biometric authentication (2020-11-10),
14. 10810487 - Reconfigurable and customizable general-purpose circuits for neural networks (2020-10-20),
15. 10706923 - Resistive random-access memory for exclusive NOR (XNOR) neural networks (2020-07-07),

Title: The Innovative Contributions of Jae-sun Seo in Computing Technology

Introduction: Jae-sun Seo, based in Tempe, Arizona, is an accomplished inventor holding an impressive 31 patents. His groundbreaking work primarily focuses on in-memory computing and enhancing the efficiency of convolutional neural networks. His contributions have significantly advanced the field of computing technology, making processes more efficient and effective.

Latest Patents: Among his latest innovations, Jae-sun has developed patents for "Circuits and Methods for In-Memory Computing." This innovative SRAM macro, known as 'C3SRAM,' utilizes capacitive-coupling computing to support fully parallel computation and configurable outputs. It incorporates circuits within bitcells to accelerate hardware for neural networks, particularly with binarized weights and activations. The technology achieves fully parallel vector-matrix multiplication in a single cycle, enhancing speed and efficiency.

Another notable patent is "Cascaded Computing for Convolutional Neural Networks," which focuses on reducing computation in CNNs without sacrificing output accuracy. By exploiting convolutional and pooling operations, Jae-sun's techniques minimize computation redundancy, allowing for a maximum to be identified using a significantly smaller number of bits. This approach ensures efficient performance while maintaining the integrity of the final classification results.

Career Highlights: Jae-sun Seo has held significant positions at esteemed organizations such as IBM and Arizona State University. His work at these institutions has been pivotal in shaping advancements in computing technologies. He has made substantial contributions to the industry through his innovative patents and research initiatives, reflecting his dedication to driving technological progress.

Collaborations: Throughout his career, Jae-sun has collaborated with notable professionals such as Leland Chang and Robert K. Montoye. These partnerships have facilitated the development of groundbreaking technologies and have significantly influenced the research landscape within the field of computing.

Conclusion: Jae-sun Seo's contributions to the realm of computing are marked by his extensive patent portfolio and impactful innovations. His work not only advances technological capabilities but also sets a foundation for future developments in the field. As he continues to explore new horizons, the impact of his inventions will probably resonate within the industry for years to come.

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…