Restricting the Flow: Information Bottlenecks for Attribution

TL;DR

This paper introduces an information bottleneck approach for attribution in neural networks, quantifying the importance of input regions in bits and outperforming existing methods across multiple metrics.

Contribution

It adapts the information bottleneck concept for attribution, providing an information-theoretic measure and guarantees for input region importance in neural network decisions.

Findings

Outperforms ten baselines in most settings

Provides an absolute information measure in bits

Guarantees non-essential regions have near-zero relevance

Abstract

Attribution methods provide insights into the decision-making of machine learning models like artificial neural networks. For a given input sample, they assign a relevance score to each individual input variable, such as the pixels of an image. In this work we adapt the information bottleneck concept for attribution. By adding noise to intermediate feature maps we restrict the flow of information and can quantify (in bits) how much information image regions provide. We compare our method against ten baselines using three different metrics on VGG-16 and ResNet-50, and find that our methods outperform all baselines in five out of six settings. The method's information-theoretic foundation provides an absolute frame of reference for attribution values (bits) and a guarantee that regions scored close to zero are not necessary for the network's decision. For reviews: https://openreview.net/forum?id=S1xWh1rYwB For code: https://github.com/BioroboticsLab/IBA