A Visual Designer of Layer-wise Relevance Propagation Models
Loading...
Date
2021
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association and John Wiley & Sons Ltd.
Abstract
Layer-wise Relevance Propagation (LRP) is an emerging and widely-used method for interpreting the prediction results of convolutional neural networks (CNN). LRP developers often select and employ different relevance backpropagation rules and parameters, to compute relevance scores on input images. However, there exists no obvious solution to define a ''best'' LRP model. A satisfied model is highly reliant on pertinent images and designers' goals. We develop a visual model designer, named as VisLRPDesigner, to overcome the challenges in the design and use of LRP models. Various LRP rules are unified into an integrated framework with an intuitive workflow of parameter setup. VisLRPDesigner thus allows users to interactively configure and compare LRP models. It also facilitates relevance-based visual analysis with two important functions: relevance-based pixel flipping and neuron ablation. Several use cases illustrate the benefits of VisLRPDesigner. The usability and limitation of the visual designer is evaluated by LRP users.
Description
@article{10.1111:cgf.14302,
journal = {Computer Graphics Forum},
title = {{A Visual Designer of Layer-wise Relevance Propagation Models}},
author = {Huang, Xinyi and Jamonnak, Suphanut and Zhao, Ye and Wu, Tsung Heng and Xu, Wei},
year = {2021},
publisher = {The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {10.1111/cgf.14302}
}