DeepAI AI Chat
Log In Sign Up

Evaluating Saliency Map Explanations for Convolutional Neural Networks: A User Study

by   Ahmed Alqaraawi, et al.
Berlin Institute of Technology (Technische Universität Berlin)

Convolutional neural networks (CNNs) offer great machine learning performance over a range of applications, but their operation is hard to interpret, even for experts. Various explanation algorithms have been proposed to address this issue, yet limited research effort has been reported concerning their user evaluation. In this paper, we report on an online between-group user study designed to evaluate the performance of "saliency maps" - a popular explanation algorithm for image classification applications of CNNs. Our results indicate that saliency maps produced by the LRP algorithm helped participants to learn about some specific image features the system is sensitive to. However, the maps seem to provide very limited help for participants to anticipate the network's output for new images. Drawing on our findings, we highlight implications for design and further research on explainable AI. In particular, we argue the HCI and AI communities should look beyond instance-level explanations.


page 1

page 4


A psychological theory of explainability

The goal of explainable Artificial Intelligence (XAI) is to generate hum...

Trustworthy Convolutional Neural Networks: A Gradient Penalized-based Approach

Convolutional neural networks (CNNs) are commonly used for image classif...

Metrics for saliency map evaluation of deep learning explanation methods

Due to the black-box nature of deep learning models, there is a recent d...

Diagrammatization: Rationalizing with diagrammatic AI explanations for abductive reasoning on hypotheses

Many visualizations have been developed for explainable AI (XAI), but th...

Easy to Decide, Hard to Agree: Reducing Disagreements Between Saliency Methods

A popular approach to unveiling the black box of neural NLP models is to...

A Theoretical Explanation for Perplexing Behaviors of Backpropagation-based Visualizations

Backpropagation-based visualizations have been proposed to interpret con...

Sanity Checks for Saliency Maps

Saliency methods have emerged as a popular tool to highlight features in...