Explaining Neural Networks Semantically and Quantitatively
arXiv:1812.07169
Abstract
This paper presents a method to explain the knowledge encoded in a convolutional neural network (CNN) quantitatively and semantically. The analysis of the specific rationale of each prediction made by the CNN presents a key issue of understanding neural networks, but it is also of significant practical values in certain applications. In this study, we propose to distill knowledge from the CNN into an explainable additive model, so that we can use the explainable model to provide a quantitative explanation for the CNN prediction. We analyze the typical bias-interpreting problem of the explainable model and develop prior losses to guide the learning of the explainable additive model. Experimental results have demonstrated the effectiveness of our method.
References in corpus (8)
- Intriguing properties of neural networks
- Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
- Born Again Neural Networks
- Distilling a Neural Network Into a Soft Decision Tree
- RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism
- Harnessing Deep Neural Networks with Logic Rules
- Explainable Neural Networks based on Additive Index Models
- Interpreting CNNs via Decision Trees