Interpretable Convolutional Neural Networks
This addresses the interpretability problem for researchers and practitioners using CNNs, though it is incremental as it builds on existing CNN architectures.
The paper tackles the problem of understanding convolutional neural networks (CNNs) by proposing a method to modify traditional CNNs into interpretable CNNs, where each filter in high conv-layers represents an object part without needing part annotations, resulting in filters that are more semantically meaningful than those in traditional CNNs.
This paper proposes a method to modify traditional convolutional neural networks (CNNs) into interpretable CNNs, in order to clarify knowledge representations in high conv-layers of CNNs. In an interpretable CNN, each filter in a high conv-layer represents a certain object part. We do not need any annotations of object parts or textures to supervise the learning process. Instead, the interpretable CNN automatically assigns each filter in a high conv-layer with an object part during the learning process. Our method can be applied to different types of CNNs with different structures. The clear knowledge representation in an interpretable CNN can help people understand the logics inside a CNN, i.e., based on which patterns the CNN makes the decision. Experiments showed that filters in an interpretable CNN were more semantically meaningful than those in traditional CNNs.