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

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 gastroenterological 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.

Contextual Importance and Utility: aTheoretical Foundation

“…The same is not true for regression tasks. For instance, in the well-known Boston Housing data set, the output value is the median value of owner-occupied homes in $1000's and is in the range [5,50]. A straightforward way of transforming that value into a utility value is an affine transformation [5,50] → [0, 1], assuming that the preference is to have a higher value.…”

“…For instance, in the well-known Boston Housing data set, the output value is the median value of owner-occupied homes in $1000's and is in the range [5,50]. A straightforward way of transforming that value into a utility value is an affine transformation [5,50] → [0, 1], assuming that the preference is to have a higher value. However, from a buyer's point of view, the preference might be for lower prices and then the transformation would rather be [50, 5] → [0, 1].…”

“…The CIU Github site https://github.com/ KaryFramling/ciu provides executable examples at least for the well-known benchmark data sets Iris, Boston, Heart Disease, UCI Cars, Diamonds, Titanic and Adult and several different machine learning models. CIU is also implemented for image explanations as reported in [5,7]. The source code used in this paper is published at https://github.com/KaryFramling/AJCAI_2021.…”