18227261. METHOD OF LEARNING NEURAL NETWORK, FEATURE SELECTION APPARATUS, FEATURE SELECTION METHOD, AND RECORDING MEDIUM simplified abstract (NEC Corporation)

From WikiPatents
Jump to navigation Jump to search

METHOD OF LEARNING NEURAL NETWORK, FEATURE SELECTION APPARATUS, FEATURE SELECTION METHOD, AND RECORDING MEDIUM

Organization Name

NEC Corporation

Inventor(s)

Masanao Natsumeda of Tokyo (JP)

METHOD OF LEARNING NEURAL NETWORK, FEATURE SELECTION APPARATUS, FEATURE SELECTION METHOD, AND RECORDING MEDIUM - A simplified explanation of the abstract

This abstract first appeared for US patent application 18227261 titled 'METHOD OF LEARNING NEURAL NETWORK, FEATURE SELECTION APPARATUS, FEATURE SELECTION METHOD, AND RECORDING MEDIUM

Simplified Explanation

The abstract describes a method for learning a neural network that includes a feature selection layer, a feature extraction layer, and a prediction layer. The method involves adjusting the weight parameter of the neural network to improve prediction accuracy and reduce the contribution of the input data domain to the prediction result.

  • The neural network includes a feature selection layer, feature extraction layer, and prediction layer.
  • The feature selection layer selects a portion of the input data that contains information about the domain of each sample.
  • The feature extraction layer extracts a feature quantity based on the selected input data.
  • The prediction layer performs predictions based on the feature quantity.
  • The weight parameter of the neural network is adjusted to increase prediction accuracy.
  • The weight parameter is also adjusted to reduce the influence of the input data domain on the prediction result.

Potential Applications:

  • This method can be applied in various domains where accurate predictions are required, such as finance, healthcare, and weather forecasting.
  • It can be used in image recognition systems to improve accuracy by considering the domain information of the input images.

Problems Solved:

  • The method addresses the problem of low prediction accuracy in neural networks by adjusting the weight parameter.
  • It solves the issue of the input data domain having a significant impact on the prediction result by reducing its contribution.

Benefits:

  • Improved prediction accuracy leads to more reliable and accurate results in various applications.
  • Reducing the influence of the input data domain allows for more generalizable predictions that are not overly biased by specific domains.


Original Abstract Submitted

A method of learning a neural network, wherein the neural network includes: a feature selection layer for selecting a part of input data including information about a domain of each sample; a feature extraction layer for extracting a feature quantity on the basis of the selected input data; and a prediction layer for performing a prediction on the basis of the feature quantity, and the method includes adjusting a weight parameter of the neural network to increase a prediction accuracy by the prediction layer and to reduce a contribution to a prediction result of the prediction layer by the domain of the input data.

Retrieved from "http://wikipatents.org/index.php?title=18227261._METHOD_OF_LEARNING_NEURAL_NETWORK,_FEATURE_SELECTION_APPARATUS,_FEATURE_SELECTION_METHOD,_AND_RECORDING_MEDIUM_simplified_abstract_(NEC_Corporation)&oldid=31872"