Skip to main content

ciu.image: An R Package for Explaining Image Classification with Contextual Importance and Utility.

Främling, K., Knapic̆, S. and Malhi, A., 2021. ciu.image: An R Package for Explaining Image Classification with Contextual Importance and Utility. In: EXTRAAMAS 2021: Third International Workshop on Explainable, Transparent Autonomous Agents and Multi-Agent Systems, 3-7 May 2021, Virtual, 55 - 62.

Full text available as:

[img]
Preview
PDF
EXTRAAMAS_2021_CIU_image.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.

2MB

DOI: 10.1007/978-3-030-82017-6_4

Abstract

Many techniques have been proposed in recent years that attempt to explain results of image classifiers, notably for the case when the classifier is a deep neural network. This paper presents an implementation of the Contextual Importance and Utility method for explaining image classifications. It is an R package that can be used with the most usual image classification models. The paper shows results for typical benchmark images, as well as for a medical data set of gastro-enterological images. For comparison, results produced by the LIME method are included. Results show that CIU produces similar or better results than LIME with significantly shorter calculation times. However, the main purpose of this paper is to bring the existence of this package to general knowledge and use, rather than comparing with other explanation methods.

Item Type:Conference or Workshop Item (Paper)
ISSN:0302-9743
Additional Information:The work is partially supported by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation.
Uncontrolled Keywords:Explainable Artificial Intelligence; Contextual Importance and Utility; Image Classification; Deep Neural Network
Group:Faculty of Science & Technology
ID Code:36357
Deposited By: Unnamed user with email symplectic@symplectic
Deposited On:13 Dec 2021 10:48
Last Modified:13 Dec 2021 10:48

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -