1. Interpretable Compositional Convolutional Neural Networks
- Author
-
Jiaqi Fan, Shikun Huang, Zhihua Wei, Binbin Zhang, Quanshi Zhang, Ping Zhao, and Wen Shen
- Subjects
FOS: Computer and information sciences ,Computer science ,business.industry ,Computer Vision and Pattern Recognition (cs.CV) ,Computer Science - Computer Vision and Pattern Recognition ,Pattern recognition ,Filter (signal processing) ,ENCODE ,Object (computer science) ,Convolutional neural network ,Image (mathematics) ,Visual patterns ,Core (graph theory) ,Artificial intelligence ,business ,Interpretability - Abstract
The reasonable definition of semantic interpretability presents the core challenge in explainable AI. This paper proposes a method to modify a traditional convolutional neural network (CNN) into an interpretable compositional CNN, in order to learn filters that encode meaningful visual patterns in intermediate convolutional layers. In a compositional CNN, each filter is supposed to consistently represent a specific compositional object part or image region with a clear meaning. The compositional CNN learns from image labels for classification without any annotations of parts or regions for supervision. Our method can be broadly applied to different types of CNNs. Experiments have demonstrated the effectiveness of our method., Comment: IJCAI2021
- Published
- 2021
- Full Text
- View/download PDF