A multiple spiking neural network architecture based on fuzzy intervals for anomaly detection: a case study of rail defects

W. Phusakulkajorn; J.M. Hendriks; J. Moraal; R.P.B.J. Dollevoet; Z. Li; Alfredo Nunez

doi:10.1109/FUZZ-IEEE55066.2022.9882864

A multiple spiking neural network architecture based on fuzzy intervals for anomaly detection: a case study of rail defects

W. Phusakulkajorn, J.M. Hendriks, J. Moraal, R.P.B.J. Dollevoet, Z. Li, Alfredo Nunez

Railway Engineering

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

1 Citation (Scopus)

30 Downloads (Pure)

Abstract

In this paper, a fuzzy interval-based method is proposed for solving the problem of rail defect detection relying on an on-board measurement system and a multiple spiking neural network architecture. Instead of outputting binary values (defect or not defect), all data will belong to both classes with different spreads that are given by two fuzzy intervals. The multiple spiking neural networks are used to capture different sources of uncertainties. In this paper, we consider uncertainties in the parameters of spiking neural networks during the training phase. The proposed method comprises two steps. In the first step,
multiple sets of the firing times for both classes are obtained from multiple spiking neural networks. In the second step, the obtained multiple sets of firing times are fuzzy numbers and they are used to construct fuzzy intervals. The proposed method is showcased with the problem of rail defect detection. The
numerical analysis indicates that the fuzzy intervals are suitable to make use of the information provided by the multiple spike neural networks. Finally, with the proposed method, we improve the interpretability of the decision making regarding the detection of anomalies.

Original language	English
Title of host publication	2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Number of pages	8
Volume	2022
ISBN (Electronic)	978-1-6654-6710-0
DOIs	https://doi.org/10.1109/FUZZ-IEEE55066.2022.9882864
Publication status	Published - 2022
Event	2022 IEEE International Conference on Fuzzy Systems - Centro Congressi Padova, Padua, Italy Duration: 18 Jul 2022 → 23 Jul 2022 https://wcci2022.org/

Conference

Conference	2022 IEEE International Conference on Fuzzy Systems
Abbreviated title	FUZZ-IEEE 2022
Country/Territory	Italy
City	Padua
Period	18/07/22 → 23/07/22
Internet address	https://wcci2022.org/

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

spiking neural network
parameter uncertainty
prediction interval
interpretability
anomaly detection

Access to Document

10.1109/FUZZ-IEEE55066.2022.9882864

A_multiple_spiking_neural_network_architecture_based_on_fuzzy_intervals_for_anomaly_detection_a_case_study_of_rail_defectsFinal published version, 14.2 MB

Cite this

Phusakulkajorn, W., Hendriks, J. M., Moraal, J., Dollevoet, R. P. B. J., Li, Z., & Nunez, A. (2022). A multiple spiking neural network architecture based on fuzzy intervals for anomaly detection: a case study of rail defects. In 2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) (Vol. 2022). Article 1292 Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/FUZZ-IEEE55066.2022.9882864

@inproceedings{2d80cc65d7ea4a5fa2e0191acd31c708,

title = "A multiple spiking neural network architecture based on fuzzy intervals for anomaly detection: a case study of rail defects",

abstract = "In this paper, a fuzzy interval-based method is proposed for solving the problem of rail defect detection relying on an on-board measurement system and a multiple spiking neural network architecture. Instead of outputting binary values (defect or not defect), all data will belong to both classes with different spreads that are given by two fuzzy intervals. The multiple spiking neural networks are used to capture different sources of uncertainties. In this paper, we consider uncertainties in the parameters of spiking neural networks during the training phase. The proposed method comprises two steps. In the first step,multiple sets of the firing times for both classes are obtained from multiple spiking neural networks. In the second step, the obtained multiple sets of firing times are fuzzy numbers and they are used to construct fuzzy intervals. The proposed method is showcased with the problem of rail defect detection. Thenumerical analysis indicates that the fuzzy intervals are suitable to make use of the information provided by the multiple spike neural networks. Finally, with the proposed method, we improve the interpretability of the decision making regarding the detection of anomalies.",

keywords = "spiking neural network, parameter uncertainty, prediction interval, interpretability, anomaly detection",

author = "W. Phusakulkajorn and J.M. Hendriks and J. Moraal and R.P.B.J. Dollevoet and Z. Li and Alfredo Nunez",

note = "Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; 2022 IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2022 ; Conference date: 18-07-2022 Through 23-07-2022",

year = "2022",

doi = "10.1109/FUZZ-IEEE55066.2022.9882864",

language = "English",

volume = "2022",

booktitle = "2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

url = "https://wcci2022.org/",

}

Phusakulkajorn, W , Hendriks, JM , Moraal, J , Dollevoet, RPBJ , Li, Z & Nunez, A 2022, A multiple spiking neural network architecture based on fuzzy intervals for anomaly detection: a case study of rail defects. in 2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE). vol. 2022, 1292, Institute of Electrical and Electronics Engineers (IEEE), 2022 IEEE International Conference on Fuzzy Systems, Padua, Italy, 18/07/22. https://doi.org/10.1109/FUZZ-IEEE55066.2022.9882864

A multiple spiking neural network architecture based on fuzzy intervals for anomaly detection: a case study of rail defects. / Phusakulkajorn, W.; Hendriks, J.M.; Moraal, J. et al.
2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE). Vol. 2022 Institute of Electrical and Electronics Engineers (IEEE), 2022. 1292.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - A multiple spiking neural network architecture based on fuzzy intervals for anomaly detection

T2 - 2022 IEEE International Conference on Fuzzy Systems

AU - Phusakulkajorn, W.

AU - Hendriks, J.M.

AU - Moraal, J.

AU - Dollevoet, R.P.B.J.

AU - Li, Z.

AU - Nunez, Alfredo

N1 - Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2022

Y1 - 2022

N2 - In this paper, a fuzzy interval-based method is proposed for solving the problem of rail defect detection relying on an on-board measurement system and a multiple spiking neural network architecture. Instead of outputting binary values (defect or not defect), all data will belong to both classes with different spreads that are given by two fuzzy intervals. The multiple spiking neural networks are used to capture different sources of uncertainties. In this paper, we consider uncertainties in the parameters of spiking neural networks during the training phase. The proposed method comprises two steps. In the first step,multiple sets of the firing times for both classes are obtained from multiple spiking neural networks. In the second step, the obtained multiple sets of firing times are fuzzy numbers and they are used to construct fuzzy intervals. The proposed method is showcased with the problem of rail defect detection. Thenumerical analysis indicates that the fuzzy intervals are suitable to make use of the information provided by the multiple spike neural networks. Finally, with the proposed method, we improve the interpretability of the decision making regarding the detection of anomalies.

AB - In this paper, a fuzzy interval-based method is proposed for solving the problem of rail defect detection relying on an on-board measurement system and a multiple spiking neural network architecture. Instead of outputting binary values (defect or not defect), all data will belong to both classes with different spreads that are given by two fuzzy intervals. The multiple spiking neural networks are used to capture different sources of uncertainties. In this paper, we consider uncertainties in the parameters of spiking neural networks during the training phase. The proposed method comprises two steps. In the first step,multiple sets of the firing times for both classes are obtained from multiple spiking neural networks. In the second step, the obtained multiple sets of firing times are fuzzy numbers and they are used to construct fuzzy intervals. The proposed method is showcased with the problem of rail defect detection. Thenumerical analysis indicates that the fuzzy intervals are suitable to make use of the information provided by the multiple spike neural networks. Finally, with the proposed method, we improve the interpretability of the decision making regarding the detection of anomalies.

KW - spiking neural network

KW - parameter uncertainty

KW - prediction interval

KW - interpretability

KW - anomaly detection

UR - http://www.scopus.com/inward/record.url?scp=85138816090&partnerID=8YFLogxK

U2 - 10.1109/FUZZ-IEEE55066.2022.9882864

DO - 10.1109/FUZZ-IEEE55066.2022.9882864

M3 - Conference contribution

VL - 2022

BT - 2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)

PB - Institute of Electrical and Electronics Engineers (IEEE)

Y2 - 18 July 2022 through 23 July 2022

ER -

Phusakulkajorn W , Hendriks JM , Moraal J , Dollevoet RPBJ , Li Z , Nunez A. A multiple spiking neural network architecture based on fuzzy intervals for anomaly detection: a case study of rail defects. In 2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE). Vol. 2022. Institute of Electrical and Electronics Engineers (IEEE). 2022. 1292 doi: 10.1109/FUZZ-IEEE55066.2022.9882864

A multiple spiking neural network architecture based on fuzzy intervals for anomaly detection: a case study of rail defects

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this