Skip to content
/ CCNN Public

Comprehensible Convolutional Neural Networks via Guided Concept Learning

Notifications You must be signed in to change notification settings

sandareka/CCNN

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Comprehensible Convolutional Neural Network via Guided Concept Learning

Official implementation of Comprehensible Convolutional Neural Network via Guided Concept Learning accepted to IJCNN 2021

Abstract

Learning concepts that are consistent with human perception is important for Deep Neural Networks to win end-user trust. Post-hoc interpretation methods lack transparency in the feature representations learned by the models. This work proposes a guided learning approach with an additional concept layer in a CNN-based architecture to learn the associations between visual features and word phrases. We design an objective function that optimizes both prediction accuracy and semantics of the learned feature representations. Experiment results demonstrate that the proposed model can learn concepts that are consistent with human perception and their corresponding contributions to the model decision without compromising accuracy. Further, these learned concepts are transferable to new classes of objects that have similar concepts.

Presentation Video : https://www.youtube.com/watch?v=vK4vti_pUMg

Overview of Comprehensible CNN

ccnnOverview

Citation

@inproceedings{wickramanayake2021comprehensible,
  title={Comprehensible convolutional neural networks via guided concept learning},
  author={Wickramanayake, Sandareka and Hsu, Wynne and Lee, Mong Li},
  booktitle={2021 International Joint Conference on Neural Networks (IJCNN)},
  pages={1--8},
  year={2021},
  organization={IEEE}
}