Standard

Explainable AI: Graph Based Sampling Approach for High Dimensional AI System. / Zou, Jinying; Xu, Feiran; Petrosian, Ovanes; Li, Yin.

в: Lecture Notes in Networks and Systems, № 776, 21.09.2023, стр. 410-422.

Результаты исследований: Научные публикации в периодических изданиях › статья › Рецензирование

Harvard

Zou, J, Xu, F, Petrosian, O & Li, Y 2023, 'Explainable AI: Graph Based Sampling Approach for High Dimensional AI System', Lecture Notes in Networks and Systems, № 776, стр. 410-422. https://doi.org/10.1007/978-3-031-43789-2_38

APA

Zou, J., Xu, F., Petrosian, O., & Li, Y. (2023). Explainable AI: Graph Based Sampling Approach for High Dimensional AI System. Lecture Notes in Networks and Systems, ( 776), 410-422. https://doi.org/10.1007/978-3-031-43789-2_38

Vancouver

Zou J, Xu F, Petrosian O, Li Y. Explainable AI: Graph Based Sampling Approach for High Dimensional AI System. Lecture Notes in Networks and Systems. 2023 Сент. 21;( 776):410-422. https://doi.org/10.1007/978-3-031-43789-2_38

Author

Zou, Jinying ; Xu, Feiran ; Petrosian, Ovanes ; Li, Yin. / Explainable AI: Graph Based Sampling Approach for High Dimensional AI System. в: Lecture Notes in Networks and Systems. 2023 ; № 776. стр. 410-422.

BibTeX

@article{8bc4205684fe410785c6187fad3ac487,
title = "Explainable AI: Graph Based Sampling Approach for High Dimensional AI System",
abstract = "The widespread use of AI in various industries has been facilitated by advancements in machine learning and neural networks. To shed light on the workings of opaque data-driven algorithms, several mathematical methods have emerged, such as the Shapley value, tree models, and Taylor expansion. Among these, the Shapley value stands out as a popular perturbation method, garnering significant attention. While calculating Shapley values is known to be an NP-hard problem, some researchers have introduced approximate techniques to alleviate this challenge. However, striking a balance between accuracy and time cost remains difficult, particularly as the number of players involved increases. In this paper, we propose a novel approach that efficiently computes Shapley values using fewer high-quality coalition samples, relying on the relationship map.",
keywords = "Explainable AI, Anomaly Detection, High-dimension Problem, Graph Based Sampling, Shapley Value, Anomaly Detection, Explainable AI, Graph Based Sampling, High-dimension Problem, Shapley Value",
author = "Jinying Zou and Feiran Xu and Ovanes Petrosian and Yin Li",
year = "2023",
month = sep,
day = "21",
doi = "10.1007/978-3-031-43789-2_38",
language = "English",
pages = "410--422",
journal = "Lecture Notes in Networks and Systems",
issn = "2367-3389",
publisher = "Springer Nature",
number = " 776",

}

RIS

TY - JOUR

T1 - Explainable AI: Graph Based Sampling Approach for High Dimensional AI System

AU - Zou, Jinying

AU - Xu, Feiran

AU - Petrosian, Ovanes

AU - Li, Yin

PY - 2023/9/21

Y1 - 2023/9/21

N2 - The widespread use of AI in various industries has been facilitated by advancements in machine learning and neural networks. To shed light on the workings of opaque data-driven algorithms, several mathematical methods have emerged, such as the Shapley value, tree models, and Taylor expansion. Among these, the Shapley value stands out as a popular perturbation method, garnering significant attention. While calculating Shapley values is known to be an NP-hard problem, some researchers have introduced approximate techniques to alleviate this challenge. However, striking a balance between accuracy and time cost remains difficult, particularly as the number of players involved increases. In this paper, we propose a novel approach that efficiently computes Shapley values using fewer high-quality coalition samples, relying on the relationship map.

AB - The widespread use of AI in various industries has been facilitated by advancements in machine learning and neural networks. To shed light on the workings of opaque data-driven algorithms, several mathematical methods have emerged, such as the Shapley value, tree models, and Taylor expansion. Among these, the Shapley value stands out as a popular perturbation method, garnering significant attention. While calculating Shapley values is known to be an NP-hard problem, some researchers have introduced approximate techniques to alleviate this challenge. However, striking a balance between accuracy and time cost remains difficult, particularly as the number of players involved increases. In this paper, we propose a novel approach that efficiently computes Shapley values using fewer high-quality coalition samples, relying on the relationship map.

KW - Explainable AI

KW - Anomaly Detection

KW - High-dimension Problem

KW - Graph Based Sampling

KW - Shapley Value

KW - Anomaly Detection

KW - Explainable AI

KW - Graph Based Sampling

KW - High-dimension Problem

KW - Shapley Value

UR - https://www.mendeley.com/catalogue/80e2515e-0c87-3509-8b9f-7f7f937698da/

U2 - 10.1007/978-3-031-43789-2_38

DO - 10.1007/978-3-031-43789-2_38

M3 - Article

SP - 410

EP - 422

JO - Lecture Notes in Networks and Systems

JF - Lecture Notes in Networks and Systems

SN - 2367-3389

IS - 776

ER -

ID: 114434352