Contextual Semantic Interpretability

TL;DR

This paper introduces a contextual semantic bottleneck in CNNs that groups interpretable attributes to improve model transparency while maintaining high prediction accuracy in landscape scenicness estimation.

Contribution

It proposes a novel semantic bottleneck that captures context by grouping attributes, enhancing interpretability without sacrificing performance.

Findings

Achieves comparable accuracy to non-interpretable models on scenicness estimation.

Provides clear, interpretable explanations for each prediction.

Demonstrates effective use of semantic groups in CNN interpretability.

Abstract

Convolutional neural networks (CNN) are known to learn an image representation that captures concepts relevant to the task, but do so in an implicit way that hampers model interpretability. However, one could argue that such a representation is hidden in the neurons and can be made explicit by teaching the model to recognize semantically interpretable attributes that are present in the scene. We call such an intermediate layer a \emph{semantic bottleneck}. Once the attributes are learned, they can be re-combined to reach the final decision and provide both an accurate prediction and an explicit reasoning behind the CNN decision. In this paper, we look into semantic bottlenecks that capture context: we want attributes to be in groups of a few meaningful elements and participate jointly to the final decision. We use a two-layer semantic bottleneck that gathers attributes into interpretable, sparse groups, allowing them contribute differently to the final output depending on the context. We test our contextual semantic interpretable bottleneck (CSIB) on the task of landscape scenicness estimation and train the semantic interpretable bottleneck using an auxiliary database (SUN Attributes). Our model yields in predictions as accurate as a non-interpretable baseline when applied to a real-world test set of Flickr images, all while providing clear and interpretable explanations for each prediction.