TY - GEN

T1 - XPROAX

T2 - 8th IEEE International Conference on Data Science and Advanced Analytics, DSAA 2021

AU - Cai, Yi

AU - Zimek, Arthur

AU - Ntoutsi, Eirini

N1 - Funding Information: The first author is supported by the State Ministry of Science and Culture of Lower Saxony, within the PhD program “Responsible Artificial Intelligence in the Digital Society”. We also thank Philip Naumann for the insightful discussions.

PY - 2021

Y1 - 2021

N2 - 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. To overcome this problem, we propose a progressive approximation of the neighborhood using counterfactual instances as initial landmarks and a careful 2-stage sampling approach to refine counterfactuals and generate factuals in the neighborhood of the input instance to be explained. Our work focuses on textual data and our explanations consist of both word-level explanations from the original instance (intrinsic) and the neighborhood (extrinsic) and factual- and counterfactual-instances discovered during the neighborhood generation process that further reveal the effect of altering certain parts in the input text. Our experiments on real-world datasets demonstrate that our method outperforms the competitors in terms of usefulness and stability (for the qualitative part) and completeness, compactness and correctness (for the quantitative part).

AB - 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. To overcome this problem, we propose a progressive approximation of the neighborhood using counterfactual instances as initial landmarks and a careful 2-stage sampling approach to refine counterfactuals and generate factuals in the neighborhood of the input instance to be explained. Our work focuses on textual data and our explanations consist of both word-level explanations from the original instance (intrinsic) and the neighborhood (extrinsic) and factual- and counterfactual-instances discovered during the neighborhood generation process that further reveal the effect of altering certain parts in the input text. Our experiments on real-world datasets demonstrate that our method outperforms the competitors in terms of usefulness and stability (for the qualitative part) and completeness, compactness and correctness (for the quantitative part).

KW - Counterfactuals

KW - Explainable AI

KW - Local explanations

KW - Neighborhood approximation

KW - Text classification

UR - http://www.scopus.com/inward/record.url?scp=85126082401&partnerID=8YFLogxK

U2 - 10.48550/arXiv.2109.15004

DO - 10.48550/arXiv.2109.15004

M3 - Conference contribution

AN - SCOPUS:85126082401

SN - 978-1-6654-2100-3

T3 - 2021 IEEE 8th International Conference on Data Science and Advanced Analytics, DSAA 2021

BT - 2021 IEEE 8th International Conference on Data Science and Advanced Analytics, DSAA 2021

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 6 October 2021 through 9 October 2021

ER -