Springe direkt zu Inhalt

XPROAX-Local explanations for text classification with progressive neighborhood approximation

Prof. Dr. Eirini Ntoutsi, Prof. Dr. Arthur Zimek, Yi Cai – 2021

The importance of the neighborhood for training a local surrogate model to approximate the local decision boundary of a black box classifier has been already highlighted in the literature. Several attempts have been made to construct a better neighborhood for high dimensional data, like texts, by using generative autoencoders. However, existing approaches mainly generate neighbors by selecting purely at random from the latent space and struggle under the curse of dimensionality to learn a good local decision boundary.

Titel
XPROAX-Local explanations for text classification with progressive neighborhood approximation
Verfasser
Prof. Dr. Eirini Ntoutsi, Prof. Dr. Arthur Zimek, Yi Cai
Schlagwörter
Explainable AI, Local explanations, Counterfactuals, Neighborhood approximation, Text classification
Datum
2021-07
Erschienen in
The 8th IEEE International Conference on Data Science and Advanced Analytics (DSAA) (to appear)
Sprache
eng
Art
Text
Größe oder Länge
10 pages