Nec corporation (20240160947). LEARNING DEVICE, LEARNING METHOD, AND STORAGE MEDIUM simplified abstract
Contents
- 1 LEARNING DEVICE, LEARNING METHOD, AND STORAGE MEDIUM
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 LEARNING DEVICE, LEARNING METHOD, AND STORAGE MEDIUM - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Unanswered Questions
- 1.11 Original Abstract Submitted
LEARNING DEVICE, LEARNING METHOD, AND STORAGE MEDIUM
Organization Name
Inventor(s)
LEARNING DEVICE, LEARNING METHOD, AND STORAGE MEDIUM - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240160947 titled 'LEARNING DEVICE, LEARNING METHOD, AND STORAGE MEDIUM
Simplified Explanation
The learning device for a neural network described in the abstract uses a base data group to update parameter values of the network and normalization layers, as well as adversarial examples to induce errors in estimation and update parameter values accordingly.
- Neural network learning device:
* Utilizes base data group for parameter updates * Incorporates adversarial examples for error induction and parameter updates * Includes partial network, first normalization layer, and second normalization layer
Potential Applications
The technology could be applied in:
- Cybersecurity for detecting and preventing adversarial attacks
- Image recognition for improving accuracy and robustness of neural networks
Problems Solved
This technology addresses:
- Vulnerability to adversarial attacks in neural networks
- Improving the accuracy and reliability of neural network estimations
Benefits
The benefits of this technology include:
- Enhanced security against adversarial attacks
- Improved performance and reliability of neural network estimations
Potential Commercial Applications
- Cybersecurity companies could integrate this technology into their systems
- Tech companies developing image recognition software could benefit from implementing this innovation
Possible Prior Art
One possible prior art could be the use of adversarial examples in neural networks to improve robustness and accuracy. Another could be the utilization of normalization layers for data processing and optimization.
Unanswered Questions
How does the learning device handle different types of adversarial examples?
The article does not specify how the device distinguishes between various adversarial examples and their impact on parameter updates.
What is the computational cost of implementing this learning device?
The article does not mention the computational resources required to integrate this technology into existing neural network systems.
Original Abstract Submitted
a learning device for a neural network uses a base data group, which is a group including a plurality of data, to update a parameter value of the partial network and a parameter value of the second normalization layer, and uses an adversarial example determined to induce an error in estimation using the neural network, among adversarial examples included in an adversarial data group, which is a group including a plurality of adversarial examples with respect to the data included in the base data group, to update the parameter value of the partial network and a parameter value of the first normalization layer. the neural network includes a partial network, a first normalization layer normalizing data input to the first normalization layer itself, and a second normalization layer normalizing data input to the second normalization layer itself.
