Unlock instant, AI-driven research and patent intelligence for your innovation.

Domain adaptation method and device for neural network and storage medium

A neural network, self-adaptive technology, applied in deep neural network, supervised field, can solve problems such as self-adaptation in inapplicable fields

Pending Publication Date: 2022-05-20
FUJITSU LTD
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these existing data augmentation methods all perform data augmentation on a single domain and are not suitable for domain adaptation

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Domain adaptation method and device for neural network and storage medium
  • Domain adaptation method and device for neural network and storage medium
  • Domain adaptation method and device for neural network and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] Exemplary embodiments of the present disclosure will be described below with reference to the accompanying drawings. In the interest of clarity and conciseness, not all features of an actual implementation are described in this specification. It should be understood, however, that in developing any such practical implementation, many implementation-specific decisions must be made in order to achieve the developer's specific goals, such as meeting those system- and business-related constraints and those Restrictions may vary from implementation to implementation. Moreover, it should also be understood that development work, while potentially complex and time-consuming, would at least be a routine undertaking for those skilled in the art having the benefit of this disclosure.

[0037] Here, it should be noted that in order to avoid obscuring the present disclosure due to unnecessary details, only the device structure and / or processing steps closely related to the solutio...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a domain adaptation method and device for a neural network and a computer readable storage medium. The method comprises the steps that a first sample and a second sample from a source domain are mixed with a third sample from a target domain to obtain a mixed sample, and the first sample, the second sample, the third sample and the mixed sample belong to the same category; constructing a first loss function based on the sum of the intra-class distance and the inter-class distance between the sample in the source domain and the mixed sample, wherein the first loss function minimizes the intra-class distance after weighted averaging and maximizes the inter-class distance after weighted averaging; based on the sum of the first loss function and the cross entropy loss function, constructing a second loss function for the source domain and a mixed domain composed of the mixed samples; and using both the second loss function of the source domain and the second loss function of the hybrid domain to determine whether the neural network converges, and if not, repeating the above steps.

Description

technical field [0001] The present disclosure relates to the field of deep neural networks, and in particular to supervised domain adaptation. Background technique [0002] In recent years, deep neural network models have made remarkable progress in tasks such as computer vision and natural language processing. However, these advances often rely on large-scale annotated data, such as ImageNet. When a model is directly applied to a new environment, its performance often inevitably suffers a significant drop. This is because there is a deviation between the data in the new environment and the data used for model training, that is, domain shift. Usually, the impact of domain bias can be mitigated by constructing a target dataset with enough diverse samples in the new environment, and using this dataset to retrain or fine-tune the model. However, accurate labeling of large-scale data is very expensive and time-consuming. [0003] To solve this problem, one solution is to use...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/08
CPCG06N3/08G06F18/2413G06F18/24G06F18/214
Inventor 钟朝亮汪洁冯成孙俊
Owner FUJITSU LTD