Standard

XAI Evaluation : Evaluating Black-Box Model Explanations for Prediction. / Zhang, Yuyi; Xu, Feiran; Zou, Jingying; Petrosian, Ovanes L.; Krinkin, Kirill V.

Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021. ed. / S. Shaposhnikov. Institute of Electrical and Electronics Engineers Inc., 2021. p. 13-16 9472817 (Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review

Harvard

Zhang, Y, Xu, F, Zou, J, Petrosian, OL & Krinkin, KV 2021, XAI Evaluation: Evaluating Black-Box Model Explanations for Prediction. in S Shaposhnikov (ed.), Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021., 9472817, Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021, Institute of Electrical and Electronics Engineers Inc., pp. 13-16, 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021, Saint Petersburg, Russian Federation, 16/06/21. https://doi.org/10.1109/neuront53022.2021.9472817

APA

Zhang, Y., Xu, F., Zou, J., Petrosian, O. L., & Krinkin, K. V. (2021). XAI Evaluation: Evaluating Black-Box Model Explanations for Prediction. In S. Shaposhnikov (Ed.), Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021 (pp. 13-16). [9472817] (Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/neuront53022.2021.9472817

Vancouver

Zhang Y, Xu F, Zou J, Petrosian OL, Krinkin KV. XAI Evaluation: Evaluating Black-Box Model Explanations for Prediction. In Shaposhnikov S, editor, Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 13-16. 9472817. (Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021). https://doi.org/10.1109/neuront53022.2021.9472817

Author

Zhang, Yuyi ; Xu, Feiran ; Zou, Jingying ; Petrosian, Ovanes L. ; Krinkin, Kirill V. / XAI Evaluation : Evaluating Black-Box Model Explanations for Prediction. Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021. editor / S. Shaposhnikov. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 13-16 (Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021).

BibTeX

@inproceedings{17c0f15edad64b8fb050fa9f189e156a,
title = "XAI Evaluation: Evaluating Black-Box Model Explanations for Prediction",
abstract = "The results of evaluating explanations of the black-box model for prediction are presented. The XAI evaluation is realized through the different principles and characteristics between black-box model explanations and XAI labels. In the field of high-dimensional prediction, the black-box model represented by neural network and ensemble models can predict complex data sets more accurately than traditional linear regression and white-box models such as the decision tree model. However, an unexplainable characteristic not only hinders developers from debugging but also causes users mistrust. In the XAI field dedicated to 'opening' the black box model, effective evaluation methods are still being developed. Within the established XAI evaluation framework (MDMC) in this paper, explanation methods for the prediction can be effectively tested, and the identified explanation method with relatively higher quality can improve the accuracy, transparency, and reliability of prediction. ",
keywords = "black-box model explanations, ensemble models, neural network, XAI evaluation",
author = "Yuyi Zhang and Feiran Xu and Jingying Zou and Petrosian, {Ovanes L.} and Krinkin, {Kirill V.}",
note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021 ; Conference date: 16-06-2021",
year = "2021",
month = jun,
day = "16",
doi = "10.1109/neuront53022.2021.9472817",
language = "English",
series = "Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "13--16",
editor = "S. Shaposhnikov",
booktitle = "Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021",
address = "United States",

}

RIS

TY - GEN

T1 - XAI Evaluation

T2 - 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021

AU - Zhang, Yuyi

AU - Xu, Feiran

AU - Zou, Jingying

AU - Petrosian, Ovanes L.

AU - Krinkin, Kirill V.

N1 - Publisher Copyright: © 2021 IEEE.

PY - 2021/6/16

Y1 - 2021/6/16

N2 - The results of evaluating explanations of the black-box model for prediction are presented. The XAI evaluation is realized through the different principles and characteristics between black-box model explanations and XAI labels. In the field of high-dimensional prediction, the black-box model represented by neural network and ensemble models can predict complex data sets more accurately than traditional linear regression and white-box models such as the decision tree model. However, an unexplainable characteristic not only hinders developers from debugging but also causes users mistrust. In the XAI field dedicated to 'opening' the black box model, effective evaluation methods are still being developed. Within the established XAI evaluation framework (MDMC) in this paper, explanation methods for the prediction can be effectively tested, and the identified explanation method with relatively higher quality can improve the accuracy, transparency, and reliability of prediction.

AB - The results of evaluating explanations of the black-box model for prediction are presented. The XAI evaluation is realized through the different principles and characteristics between black-box model explanations and XAI labels. In the field of high-dimensional prediction, the black-box model represented by neural network and ensemble models can predict complex data sets more accurately than traditional linear regression and white-box models such as the decision tree model. However, an unexplainable characteristic not only hinders developers from debugging but also causes users mistrust. In the XAI field dedicated to 'opening' the black box model, effective evaluation methods are still being developed. Within the established XAI evaluation framework (MDMC) in this paper, explanation methods for the prediction can be effectively tested, and the identified explanation method with relatively higher quality can improve the accuracy, transparency, and reliability of prediction.

KW - black-box model explanations

KW - ensemble models

KW - neural network

KW - XAI evaluation

UR - http://www.scopus.com/inward/record.url?scp=85112865345&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/17f506cf-4431-3e30-a462-c5b7ee8be980/

U2 - 10.1109/neuront53022.2021.9472817

DO - 10.1109/neuront53022.2021.9472817

M3 - Conference contribution

AN - SCOPUS:85112865345

T3 - Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021

SP - 13

EP - 16

BT - Proceedings of 2021 2nd International Conference on Neural Networks and Neurotechnologies, NeuroNT 2021

A2 - Shaposhnikov, S.

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 16 June 2021

ER -

ID: 86497588