Method and device for identifying classification influence factors and related equipment
A technology of influencing factors and identifying classification, applied in the field of artificial intelligence, can solve problems such as inability to effectively identify classification models
Inactive Publication Date: 2019-07-23
NEUSOFT CORP
View PDF6 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] In view of this, the embodiment of the present application provides a method, device and related equipment for identifying the influencing factors of classification, so as to solve the existing technical problem that the influencing factors of the classification model cannot be effectively identified
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0055] In order to make the above objects, features and advantages of the present application more obvious and understandable, the embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0056] In order to facilitate the understanding of the solutions provided by the embodiments of the present application, the background technology of the present application will be first described below.
[0057] In the medical field, in assisting doctors to judge patients' diseases based on deep learning models, doctors not only need to know the output results of the deep learning model, but also need to understand which factors the model produces based on the patient's output results. Therefore, visualization and explainable deep learning Modeling is an important task for increasing trust in automated medical decision-making systems. In the medical field, the factors that will have ...
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
Login to View More Abstract
The embodiment of the invention discloses a method and device for identifying classification influence factors and related equipment, and the method comprises the steps: inputting an original image into an image classification model, and obtaining a confidence value of the original image for each classification result; and sequentially shielding each pixel point in the original image to obtain a confidence value of the shielded image to each classification result; determining a relative confidence value of each pixel point to each classification result according to the confidence value of theoriginal image to each classification result and the confidence value of each pixel point to each classification result in the occlusion image. And / or determining a reverse gradient propagation valueof each pixel point according to the confidence value of the original image to each classification result and the actual classification result of the original image; and according to the relative confidence value of each pixel point to each classification result and / or the reverse gradient propagation value of each pixel point, determining an influence factor value of each pixel point to each classification result so as to determine an influence region of each classification result in the original image.
Description
technical field [0001] The present application relates to the technical field of artificial intelligence, and specifically relates to a method, device and related equipment for identifying factors affecting classification. Background technique [0002] Deep learning is a method based on data representation learning in machine learning, which has been widely used in recent years, such as image classification. The deep learning model has the characteristics of a black box, that is, by adjusting the model parameters, a deep learning model with high discrimination accuracy can be trained, but it is impossible to understand the meaning, importance, and fluctuation range of the parameters in the deep learning model through statistical basic assumptions. [0003] In fact, the deep learning model itself also means knowledge, and it is also necessary to pay attention to knowing what knowledge the deep learning model has learned from the data to produce the final decision. If a model...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/045G06F18/214G06F18/24
Inventor 王希何光宇平安
Owner NEUSOFT CORP