Do Convolutional Neural Networks Learn Class Hierarchy?

@article{Alsallakh2018DoCN,
  title={Do Convolutional Neural Networks Learn Class Hierarchy?},
  author={Bilal Alsallakh and Amin Jourabloo and Mao Ye and Xiaoming Liu and Liu Ren},
  journal={IEEE Transactions on Visualization and Computer Graphics},
  year={2018},
  volume={24},
  pages={152-162}
}
Convolutional Neural Networks (CNNs) currently achieve state-of-the-art accuracy in image classification. With a growing number of classes, the accuracy usually drops as the possibilities of confusion increase. Interestingly, the class confusion patterns follow a hierarchical structure over the classes. We present visual-analytics methods to reveal and analyze this hierarchy of similar classes in relation with CNN-internal data. We found that this hierarchy not only dictates the confusion… CONTINUE READING
Highly Cited
This paper has 39 citations. REVIEW CITATIONS
Related Discussions
This paper has been referenced on Twitter 58 times. VIEW TWEETS

Citations

Publications citing this paper.
Showing 1-10 of 21 extracted citations

Advances on Interactive Machine Learning

View 4 Excerpts
Highly Influenced

NLIZE: A Perturbation-Driven Visual Interrogation Tool for Analyzing and Interpreting Natural Language Inference Models.

IEEE transactions on visualization and computer graphics • 2018
View 4 Excerpts
Highly Influenced

Visual Analytics in Deep Learning: An Interrogative Survey for the Next Frontiers

IEEE transactions on visualization and computer graphics • 2018
View 4 Excerpts
Highly Influenced

References

Publications referenced by this paper.
Showing 1-10 of 88 references

Going deeper with convolutions

2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) • 2015
View 6 Excerpts
Highly Influenced

Visualizing Deep Convolutional Neural Networks Using Natural Pre-images

International Journal of Computer Vision • 2016
View 4 Excerpts
Highly Influenced

et al

O. Russakovsky, J. Deng, +7 authors M. Bernstein
ImageNet large scale visual recognition challenge. International Journal of Computer Vision, 115(3):211–252 • 2015
View 6 Excerpts
Highly Influenced

ImageNet: A large-scale hierarchical image database

2009 IEEE Conference on Computer Vision and Pattern Recognition • 2009
View 5 Excerpts
Highly Influenced

Towards Better Analysis of Deep Convolutional Neural Networks

IEEE Transactions on Visualization and Computer Graphics • 2017
View 3 Excerpts

Similar Papers

Loading similar papers…