Yoo Jin Choi

This inventor holds 21 USPTO granted patents and 7 published patent applications, primarily in Neural Network Optimization. Top assignees: Samsung Electronics Co., Ltd., Samsung Electronics Ltd.. Active years: 2015-2026.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Yoo Jin Choi. Retrieved from https://idiyas.com/inventor/yoo-jin-choi-ei18vp5c

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

USPTO Granted Patents = 21 (Patent Applications = 7)

% Patents Active = 90.5

Average Co-Inventor Count = 3.6

ph-index = 4

Forward Citations = 47(Granted Patents)

Company Filing History:

20 out of 132,108

Samsung Electronics Co., Ltd.

patents

1 out of 6,376

Samsung Electronics Ltd.

patents

Years Active: 2015-2026

Loading Chart...

Areas of Expertise:

Network Quantization

Computational Complexity

Convolutional Neural Networks

Adversarial Knowledge Distillation

Variable Rate Compression

Stochastic Inference

Synthetic Data Generation

Deep Learning Models

Low-Precision Neural Networks

Universal Pruning

Channel Estimation

Interference Demodulation

21 patents (USPTO):Sort by Forward CitationsExplore Patents

1. 12639581 - Method and apparatus for data-free network quantization and compression with adversarial knowledge distillation (2026-05-26),

2. 12248866 - Method and apparatus for reducing computational complexity of convolutional neural networks (2025-03-11),

3. 12190231 - Method and apparatus for neural network quantization (2025-01-07),

4. 12154030 - Method and apparatus for data-free post-training network quantization and generating synthetic data based on a pre-trained machine learning model (2024-11-26),

5. 11979175 - Method and apparatus for variable rate compression with a conditional autoencoder (2024-05-07),

6. 11699070 - Method and apparatus for providing rotational invariant neural networks (2023-07-11),

7. 11620555 - Method and apparatus for stochastic inference between multiple random variables via common representation (2023-04-04),

8. 11615317 - Method and apparatus for learning stochastic inference models between multiple random variables with unpaired data (2023-03-28),

9. 11451242 - Method and apparatus for variable rate compression with a conditional autoencoder (2022-09-20),

10. 11423312 - Method and apparatus for universal pruning and compression of deep convolutional neural networks under joint sparsity constraints (2022-08-23),

11. 11321609 - Method and apparatus for neural network quantization (2022-05-03),

12. 11270187 - Method and apparatus for learning low-precision neural network that combines weight quantization and activation quantization (2022-03-08),

13. 11164071 - Method and apparatus for reducing computational complexity of convolutional neural networks (2021-11-02),

14. 10959236 - Method and apparatus for interference demodulation reference signal detection with serving signal cancellation (2021-03-23),

15. 10886943 - Method and apparatus for variable rate compression with a conditional autoencoder (2021-01-05),

Title: Innovator Spotlight: Yoo Jin Choi - Pushing Boundaries in Image Compression and Neural Networks

Introduction:

Yoo Jin Choi, a prolific inventor based in San Diego, CA, has been making waves in the field of image compression and neural networks with a whopping 17 patents to his name. His groundbreaking work at Samsung Electronics Co., Ltd. has solidified his position as a leading figure in the industry.

Latest Patents:

1. **Method and Apparatus for Variable Rate Compression with a Conditional Autoencoder**: Choi's innovative method involves using a conditional autoencoder to achieve variable rate compression. By training the autoencoder network with different Lagrange multipliers, he has devised a system that efficiently computes compressed images based on specific schemes.

2. **Method and Apparatus for Providing Rotational Invariant Neural Networks**: In this patent, Choi introduces a unique approach to ensuring rotational invariance in neural networks. By training kernels to be symmetric, his method guarantees consistent outputs for images in different orientations.

Career Highlights:

Yoo Jin Choi's journey towards patenting 17 groundbreaking inventions has been nothing short of remarkable. His work at Samsung Electronics Co., Ltd. has not only transformed the field of image compression but has also paved the way for advancements in neural network technology. Choi's dedication to pushing the boundaries of innovation has earned him a well-deserved reputation as a visionary in the industry.

Collaborations:

Choi's success wouldn't have been possible without the support and contributions of his talented colleagues. Notable names like Jungwon Lee and Mostafa El-Khamy have worked alongside Choi, bringing their expertise to the table and collectively driving forward the innovative projects at Samsung Electronics Co., Ltd.

Conclusion:

In conclusion, Yoo Jin Choi's exceptional contributions to the realms of image compression and neural networks have undoubtedly left a lasting impact on the industry. With 17 patents under his belt and a relentless drive for innovation, Choi continues to inspire and shape the future 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]

7 patents pending: