ModelSpeX: Model Specification Using Explainable Artificial Intelligence Methods

Schlegel, Udo; Cakmak, Eren; Keim, Daniel A.

ModelSpeX: Model Specification Using Explainable Artificial Intelligence Methods

Files

007-011.pdf (405.3 KB)

Date

2020

Authors

Schlegel, Udo
Cakmak, Eren
Keim, Daniel A.

Publisher

The Eurographics Association

Abstract

Explainable artificial intelligence (XAI) methods aim to reveal the non-transparent decision-making mechanisms of black-box models. The evaluation of insight generated by such XAI methods remains challenging as the applied techniques depend on many factors (e.g., parameters and human interpretation). We propose ModelSpeX, a visual analytics workflow to interactively extract human-centered rule-sets to generate model specifications from black-box models (e.g., neural networks). The workflow enables to reason about the underlying problem, to extract decision rule sets, and to evaluate the suitability of the model for a particular task. An exemplary usage scenario walks an analyst trough the steps of the workflow to show the applicability.

        @inproceedings{10.2312:mlvis.20201100
,
booktitle = {Machine Learning Methods in Visualisation for Big Data
},
editor = {Archambault, Daniel and Nabney, Ian and Peltonen, Jaakko
},
title = {{ModelSpeX: Model Specification Using Explainable Artificial Intelligence Methods
}},
author = {Schlegel, Udo and 
Cakmak, Eren and 
Keim, Daniel A.
},
year = {2020
},
publisher = {The Eurographics Association
},
ISBN = {978-3-03868-113-7
},
DOI = {10.2312/mlvis.20201100
}
}