Where is the Model Looking At? - Concentrate and Explain the Network Attention

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

13 Scopus Citations
View graph of relations

Author(s)

  • Wenjia Xu
  • Yang Wang
  • Guangluan Xu
  • Daoyu Lin
  • Wei Dai
  • Yirong Wu

Detail(s)

Original languageEnglish
Pages (from-to)506-516
Journal / PublicationIEEE Journal on Selected Topics in Signal Processing
Volume14
Issue number3
Publication statusPublished - Mar 2020
Externally publishedYes

Abstract

Image classification models have achieved satisfactory performance on many datasets, sometimes even better than human. However, The model attention is unclear since the lack of interpretability. This paper investigates the fidelity and interpretability of model attention. We propose an Explainable Attribute-based Multi-task (EAT) framework to concentrate the model attention on the discriminative image area and make the attention interpretable. We introduce attributes prediction to the multi-task learning network, helping the network to concentrate attention on the foreground objects. We generate attribute-based textual explanations for the network and ground the attributes on the image to show visual explanations. The multi-model explanation can not only improve user trust but also help to find the weakness of network and dataset. Our framework can be generalized to any basic model. We perform experiments on three datasets and five basic models. Results indicate that the EAT framework can give multi-modal explanations that interpret the network decision. The performance of several recognition approaches is improved by guiding network attention.

Research Area(s)

  • Explainable artificial intelligence, Multi-task learning, Attributes

Citation Format(s)

Where is the Model Looking At? - Concentrate and Explain the Network Attention. / Xu, Wenjia; Wang, Jiuniu; Wang, Yang et al.
In: IEEE Journal on Selected Topics in Signal Processing, Vol. 14, No. 3, 03.2020, p. 506-516.

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review