Back to Search
Start Over
Comprehensible Convolutional Neural Networks via Guided Concept Learning
- Source :
- IJCNN
- Publication Year :
- 2021
- Publisher :
- arXiv, 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.<br />Comment: Accepted to IJCNN 2021
- Subjects :
- FOS: Computer and information sciences
Computer science
business.industry
media_common.quotation_subject
Interpretation (philosophy)
Computer Vision and Pattern Recognition (cs.CV)
Computer Science - Computer Vision and Pattern Recognition
Semantics
Convolutional neural network
Perception
Concept learning
Feature (machine learning)
Artificial intelligence
Layer (object-oriented design)
business
Word (computer architecture)
media_common
Subjects
Details
- Database :
- OpenAIRE
- Journal :
- IJCNN
- Accession number :
- edsair.doi.dedup.....43dc95608d7fb8acf9412a94d83bc4f0
- Full Text :
- https://doi.org/10.48550/arxiv.2101.03919