Global Explanations with Decision Rules: a Co-learning Approach

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférenceArticle dans les actes d'une conférence/un colloque

103 Téléchargements (Pure)


Black-box machine learning models can be extremely accurate. Yet, in critical applications such as in healthcare or justice, if models cannot be explained, domain experts will be reluctant to use them. A common way to explain a black-box model is to approximate it by a simpler model such as a decision tree. In this paper, we propose a co-learning framework to learn decision rules as explanations of black-box models through knowledge distillation and simultaneously constrain the black-box model by these explanations; all of this in a differentiable manner. To do so, we introduce the soft truncated Gaussian mixture analysis (STruGMA), a probabilistic model which encapsulates hyper-rectangle decision rules. With STruGMA, global explanations can be extracted by any rule learner such as decision lists, sets or trees. We provide evidences through experiments that our framework can globally explain differentiable black-box models such as neural networks. In particular, the explanation fidelity is increased, while the accuracy of the models is marginally impacted.
langue originaleAnglais
titreProceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence
rédacteurs en chefCassio de Campos, Marloes H. Maathuis
EditeurMLResearch Press
Nombre de pages11
Etat de la publicationPublié - 27 juil. 2021

Série de publications

NomProceedings of Machine Learning Research
ISSN (Electronique)2640-3498

Empreinte digitale

Examiner les sujets de recherche de « Global Explanations with Decision Rules: a Co-learning Approach ». Ensemble, ils forment une empreinte digitale unique.

Contient cette citation