Exclusive Feature Constrained Class Activation Mapping for Better Visual Explanation

Whereas Deep Neural Network(DNN) shows wonderful performance on large scale data, lacking interpretability limits their usage in scenarios relevant to security. To make visual explanations less noisy and more class-discriminative, in this work, we propose a visual explanation method of DNN, named Ex...

Full description

Saved in:
Bibliographic Details
Published inIEEE access Vol. 9; pp. 61417 - 61428
Main Authors Wang, Pengda, Kong, Xiangwei, Guo, Weikuo, Zhang, Xunpeng
Format Journal Article
LanguageEnglish
Published Piscataway IEEE 2021
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Artificial neural networks
class activation mapping
Constraints
Inspection
Interpretability
interpretability evaluation
Mapping
Noise measurement
Optimization
Pipelines
Predictive models
Semantics
Task analysis
Visual discrimination
visual explanation
Visualization
Online AccessGet full text

Cover

More Information
Summary:Whereas Deep Neural Network(DNN) shows wonderful performance on large scale data, lacking interpretability limits their usage in scenarios relevant to security. To make visual explanations less noisy and more class-discriminative, in this work, we propose a visual explanation method of DNN, named Exclusive Feature Constrained Class Activation Mapping(EFC-CAM). A new exclusive feature constraint is introduced to optimize the weight calculated from Grad-CAM or initialized from a constant vector. To better measure visual explanation methods, we design an effective evaluation metric which does not need bounding boxes as auxiliary information. Extensive quantitative experiments and visual inspection on ImageNet and Fashion validation set show the effectiveness of the proposed method.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2021.3073465