Enriching Visual with Verbal Explanations for Relational Concepts -- Combining LIME with Aleph

TL;DR

This paper introduces a novel method combining LIME and Aleph to generate logic-based explanations for image classifiers, capturing relational concepts and spatial relations that traditional visual explanations cannot convey.

Contribution

The approach integrates LIME with Aleph to produce relational logic rules explaining classifier decisions, enabling richer, relation-aware explanations for visual data.

Findings

Successfully identified important relational constructs in images.

Generated explicit relational rules for complex concepts like towers.

Produced explanations that relate logic rules directly to image features.

Abstract

With the increasing number of deep learning applications, there is a growing demand for explanations. Visual explanations provide information about which parts of an image are relevant for a classifier's decision. However, highlighting of image parts (e.g., an eye) cannot capture the relevance of a specific feature value for a class (e.g., that the eye is wide open). Furthermore, highlighting cannot convey whether the classification depends on the mere presence of parts or on a specific spatial relation between them. Consequently, we present an approach that is capable of explaining a classifier's decision in terms of logic rules obtained by the Inductive Logic Programming system Aleph. The examples and the background knowledge needed for Aleph are based on the explanation generation method LIME. We demonstrate our approach with images of a blocksworld domain. First, we show that our approach is capable of identifying a single relation as important explanatory construct. Afterwards, we present the more complex relational concept of towers. Finally, we show how the generated relational rules can be explicitly related with the input image, resulting in richer explanations.