Company Filing History:

12 out of 19,697

Denso Corporation

patents

2 out of 15,458

University of California

patents

where one patent can have more than one assignee

Years Active: 2018-2024

Loading Chart...

12 patents (USPTO):Sort by Forward CitationsExplore Patents

1. 12026608 - Method for adjusting output level of multilayer neural network neuron (2024-07-02),

2. 11928576 - Artificial neural network circuit and method for switching trained weight in artificial neural network circuit (2024-03-12),

3. 11586888 - Convolutional neural network (2023-02-21),

4. 11562215 - Artificial neural network circuit (2023-01-24),

5. 11537897 - Artificial neural network circuit training method, training program, and training device (2022-12-27),

6. 11501146 - Convolutional neural network (2022-11-15),

7. 11487992 - Neural network circuit (2022-11-01),

8. 11403518 - Neural network circuit (2022-08-02),

9. 11182669 - Neural network circuit (2021-11-23),

10. 10943170 - Neural network circuit (2021-03-09),

11. 10332004 - Memristive neuromorphic circuit and method for training the memristive neuromorphic circuit (2019-06-25),

12. 10074050 - Memristive neuromorphic circuit and method for training the memristive neuromorphic circuit (2018-09-11),

Title: Irina Kataeva: Innovator in Artificial Neural Networks

Introduction

Irina Kataeva is a prominent inventor based in Kariya, Japan, known for her significant contributions to the field of artificial neural networks. With a total of 12 patents to her name, she has made remarkable advancements in the technology that underpins modern machine learning and artificial intelligence.

Latest Patents

Among her latest patents is a method for adjusting the output level of a multilayer neural network neuron. This innovative method involves a multilayer neural network that includes a memristor and an analog processing circuit. The process allows for the transmission of signals between neurons and signal processing in an analog region. The adjustment step ensures that the output level of the neurons remains below the write threshold voltage of the memristor, thus optimizing the output value according to the activation function.

Another notable patent is for an artificial neural network circuit and method for switching trained weights in the circuit. This invention features a crossbar circuit that transmits signals between layered neurons, utilizing multiple memristors to assign weights to the signals. Additionally, it includes a processing circuit that calculates the sum of incoming signals while a temperature sensor detects environmental conditions, allowing for dynamic updates to the trained values used in the circuit.

Career Highlights

Irina has worked with esteemed organizations such as Denso Corporation and the University of California. Her experience in these institutions has allowed her to collaborate on cutting-edge research and development projects, further enhancing her expertise in neural network technologies.

Collaborations

Throughout her career, Irina has collaborated with notable colleagues, including Shigeki Otsuka and Dmitri Strukov. These partnerships have contributed to her innovative work and the advancement of artificial intelligence research.

Conclusion

Irina Kataeva stands out as a leading inventor in the realm of artificial neural networks, with her patents reflecting her commitment to advancing technology. Her work continues to influence the field and inspire future innovations.

This text is generated by artificial intelligence and may not be accurate.

Please report any incorrect information to support@idiyas.com

Inventor’s Patent Attorneys refers to legal professionals with specialized expertise in representing inventors throughout the patent process. These attorneys assist inventors in navigating the complexities of patent law, including filing patent applications, conducting patent searches, and protecting intellectual property rights. They play a crucial role in helping inventors secure patents for their innovative creations.