Right for the Right Reason: Making Image Classification Robust

TL;DR

This paper introduces the ObAlEx metric to evaluate if CNNs focus on the correct evidence in image classification and demonstrates that training can improve this focus without sacrificing accuracy.

Contribution

It proposes a novel explanation quality metric, ObAlEx, and shows how training can enhance CNNs' focus on relevant evidence while maintaining accuracy.

Findings

ObAlEx effectively measures focus on actual evidence.

Training improves CNN focus without accuracy loss.

Object detection aids in explanation quality assessment.

Abstract

The effectiveness of Convolutional Neural Networks (CNNs)in classifying image data has been thoroughly demonstrated. In order to explain the classification to humans, methods for visualizing classification evidence have been developed in recent years. These explanations reveal that sometimes images are classified correctly, but for the wrong reasons,i.e., based on incidental evidence. Of course, it is desirable that images are classified correctly for the right reasons, i.e., based on the actual evidence. To this end, we propose a new explanation quality metric to measure object aligned explanation in image classification which we refer to as theObAlExmetric. Using object detection approaches, explanation approaches, and ObAlEx, we quantify the focus of CNNs on the actual evidence. Moreover, we show that additional training of the CNNs can improve the focus of CNNs without decreasing their accuracy.