Interpretable Convolutional Neural Networks

CVPR 2018  ·  Quanshi Zhang, Ying Nian Wu, Song-Chun Zhu ·

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.

PDF Abstract CVPR 2018 PDF CVPR 2018 Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
single catogory classification ILSVRC Part AlexNet Accuracy 95.38 # 2
single catogory classification ILSVRC Part VGG-16 Accuracy 96.67 # 1
single catogory classification VOC Part AlexNet Accuracy 93.93 # 2
single catogory classification VOC Part VGG-16 Accuracy 95.39 # 1

Methods


No methods listed for this paper. Add relevant methods here