Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Deep learning internal data extraction method and device

A technology of data extraction and deep learning, applied in neural learning methods, reasoning methods, dynamic trees, etc., can solve problems such as lack of versatility, incorrect presentation, fuzzy deep learning analysis technology, etc., and achieve the effect of improving performance and stability

Pending Publication Date: 2022-04-12
UNIST ULSAN NAT INST OF SCI & TECH +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the large number of nodes (hidden layers) and nonlinear functions (activation functions) that make up the deep learning model, there is a problem that it is difficult to explain how data is processed and output within the deep learning model
Existing deep learning analysis techniques are too vague (e.g., may present incorrect information) to be interpreted by humans, and only exist model-specific methodologies, lacking generalizability

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
  • Deep learning internal data extraction method and device
  • Deep learning internal data extraction method and device
  • Deep learning internal data extraction method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] The specific structures or functions disclosed in this specification are merely examples to illustrate embodiments based on technical concepts, and the embodiments may have different forms, and are not limited to the embodiments in this specification.

[0040] Terms such as first or second may be used to describe various constituent elements, however, these terms are only used to distinguish one constituent element from other constituent elements. For example, a first constituent element can be named as a second constituent element, and similarly, a second constituent element can also be named as a first constituent element.

[0041] When it is described that one constituent element is "connected" or "contacted" with another constituent element, it may be directly connected or contacted with other constituent elements, but it can also be understood that there are other constituent elements therebetween. On the contrary, when it is stated that one constituent element "di...

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 data extraction method and device related to a deep learning model. According to a deep learning model of one embodiment, a data extraction method thereof comprises: a step of receiving an input query; a step of determining a first decision boundary set, the first decision boundary set being a subset of a decision boundary set corresponding to a target layer of the deep learning model; a step of extracting a decision region including the input query on the basis of the first set of decision boundaries; and a step of extracting data included in the decision region.

Description

technical field [0001] The following embodiments relate to data extraction methods and devices for deep learning models. Background technique [0002] Machine learning technology (machine learning) is an essential core technology in the era of big data (big data) or the Internet of things (IoT) that generate a large amount of data. Machine learning technology promises to have very large and wide-ranging ripple effects, as it can automate tasks in various fields that machines have not been able to do automatically. In particular, the rapid development of machine learning technology centered on deep learning has greatly narrowed the gap between the level required for practical applications and actual artificial intelligence technology, so it has attracted attention. [0003] Deep learning refers to machine learning techniques based on deep neural network models. In recent years, deep learning has played an important role in improving recognition performance in various fields...

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
IPC IPC(8): G06N3/08G06N3/04
CPCG06N3/08G06N3/04G06N3/0464G06N5/01
Inventor 崔宰植丁海东全起荣
Owner UNIST ULSAN NAT INST OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com