English

iCaps: An Interpretable Classifier via Disentangled Capsule Networks

Computer Vision and Pattern Recognition 2020-08-21 v1 Machine Learning

Abstract

We propose an interpretable Capsule Network, iCaps, for image classification. A capsule is a group of neurons nested inside each layer, and the one in the last layer is called a class capsule, which is a vector whose norm indicates a predicted probability for the class. Using the class capsule, existing Capsule Networks already provide some level of interpretability. However, there are two limitations which degrade its interpretability: 1) the class capsule also includes classification-irrelevant information, and 2) entities represented by the class capsule overlap. In this work, we address these two limitations using a novel class-supervised disentanglement algorithm and an additional regularizer, respectively. Through quantitative and qualitative evaluations on three datasets, we demonstrate that the resulting classifier, iCaps, provides a prediction along with clear rationales behind it with no performance degradation.

Keywords

Cite

@article{arxiv.2008.08756,
  title  = {iCaps: An Interpretable Classifier via Disentangled Capsule Networks},
  author = {Dahuin Jung and Jonghyun Lee and Jihun Yi and Sungroh Yoon},
  journal= {arXiv preprint arXiv:2008.08756},
  year   = {2020}
}

Comments

ECCV 2020

R2 v1 2026-06-23T17:58:45.787Z