Studying properties of water data using manifold-aware anomaly detectors

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Standard

Studying properties of water data using manifold-aware anomaly detectors. / Paulsen, Tino; Brefeld, Ulf.
Design of Cyber-Secure Water Plants. Hrsg. / Aditya Mathur; Jianying Zhou; Gauthama Raman. MDPI AG, 2024. (Water).

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Harvard

Paulsen, T & Brefeld, U 2024, Studying properties of water data using manifold-aware anomaly detectors. in A Mathur, J Zhou & G Raman (Hrsg.), Design of Cyber-Secure Water Plants. Water, MDPI AG, First International Conference on the design of cyber-secure water plants - DCS-Water'24, Buford, Georgia, USA / Vereinigte Staaten, 23.04.24.

APA

Paulsen, T., & Brefeld, U. (2024). Studying properties of water data using manifold-aware anomaly detectors. Manuskript zur Veröffentlichung eingereicht. In A. Mathur, J. Zhou, & G. Raman (Hrsg.), Design of Cyber-Secure Water Plants (Water). MDPI AG.

Vancouver

Paulsen T, Brefeld U. Studying properties of water data using manifold-aware anomaly detectors. in Mathur A, Zhou J, Raman G, Hrsg., Design of Cyber-Secure Water Plants. MDPI AG. 2024. (Water).

Bibtex

@inbook{4a8f2f9494564f1fbee82ba4f751face,
title = "Studying properties of water data using manifold-aware anomaly detectors",
abstract = "Deep learning methods, especially the family of autoencoder architectures, exhibit state-of-the-art detection rates in anomaly detection tasks. Additionally, recent results show that learning latent spaces with deep architectures on Riemannian manifolds may further improve performances as well as related interpolation tasks. In this paper, we study the use of Riemannian manifolds with variational autoencoders (VAEs) for anomaly detection on data from water providers. Besides traditional embeddings in Euclidean space, we study embeddings in Poincar{\'e} disc, spheres, and Stiefel manifolds, where in general, the Poincar{\'e} disc is often preferred for data with hierarchical structures, embeddings in spheres suggests itself for cyclical structures and the Stiefel manifold is well suited for time-dependent data. Data from water providers clearly meets all three criteria and we report on empirical results with the different manifolds as latent spaces and compare their detection performance to that of standard Euclidean embeddings. ",
keywords = "Informatics",
author = "Tino Paulsen and Ulf Brefeld",
year = "2024",
month = apr,
day = "23",
language = "English",
series = "Water",
publisher = "MDPI AG",
editor = "Aditya Mathur and Jianying Zhou and Gauthama Raman",
booktitle = "Design of Cyber-Secure Water Plants",
address = "Switzerland",
note = "First International Conference on the design of cyber-secure water plants - DCS-Water'24, DCS-Water'24 ; Conference date: 23-04-2024 Through 24-04-2024",
url = "https://itrust.sutd.edu.sg/first-international-conference-on-the-design-of-cyber-secure-water-plants-dcs-water24/paper-submission-dcs-water24/",

}

RIS

TY - CHAP

T1 - Studying properties of water data using manifold-aware anomaly detectors

AU - Paulsen, Tino

AU - Brefeld, Ulf

PY - 2024/4/23

Y1 - 2024/4/23

N2 - Deep learning methods, especially the family of autoencoder architectures, exhibit state-of-the-art detection rates in anomaly detection tasks. Additionally, recent results show that learning latent spaces with deep architectures on Riemannian manifolds may further improve performances as well as related interpolation tasks. In this paper, we study the use of Riemannian manifolds with variational autoencoders (VAEs) for anomaly detection on data from water providers. Besides traditional embeddings in Euclidean space, we study embeddings in Poincaré disc, spheres, and Stiefel manifolds, where in general, the Poincaré disc is often preferred for data with hierarchical structures, embeddings in spheres suggests itself for cyclical structures and the Stiefel manifold is well suited for time-dependent data. Data from water providers clearly meets all three criteria and we report on empirical results with the different manifolds as latent spaces and compare their detection performance to that of standard Euclidean embeddings.

AB - Deep learning methods, especially the family of autoencoder architectures, exhibit state-of-the-art detection rates in anomaly detection tasks. Additionally, recent results show that learning latent spaces with deep architectures on Riemannian manifolds may further improve performances as well as related interpolation tasks. In this paper, we study the use of Riemannian manifolds with variational autoencoders (VAEs) for anomaly detection on data from water providers. Besides traditional embeddings in Euclidean space, we study embeddings in Poincaré disc, spheres, and Stiefel manifolds, where in general, the Poincaré disc is often preferred for data with hierarchical structures, embeddings in spheres suggests itself for cyclical structures and the Stiefel manifold is well suited for time-dependent data. Data from water providers clearly meets all three criteria and we report on empirical results with the different manifolds as latent spaces and compare their detection performance to that of standard Euclidean embeddings.

KW - Informatics

UR - https://itrust.sutd.edu.sg/first-international-conference-on-the-design-of-cyber-secure-water-plants-dcs-water24/paper-submission-dcs-water24/

UR - https://www.mdpi.com/journal/water/special_issues/7UA9S2ASB0

M3 - Article in conference proceedings

T3 - Water

BT - Design of Cyber-Secure Water Plants

A2 - Mathur, Aditya

A2 - Zhou, Jianying

A2 - Raman, Gauthama

PB - MDPI AG

T2 - First International Conference on the design of cyber-secure water plants - DCS-Water'24

Y2 - 23 April 2024 through 24 April 2024

ER -

Zuletzt angesehen

Forschende

  1. Julia Perlinger

Publikationen

  1. Influences of yttrium content on microstructure and mechanical properties of as-cast Mg–Ca–Y–Zr alloys
  2. Direct measurement of cognitive load in multimedia learning
  3. Computergestütztes Repetitorium der Elementarmathematik
  4. Assessing Effects Through Laboratory Toxicity Testing
  5. Article 5 Contracts of carriage
  6. Bed-Sharing in Couples Is Associated With Increased and Stabilized REM Sleep and Sleep-Stage Synchronization
  7. An assessment of the published results of animal relocations
  8. Decentralized control of assembly processes driven by Gentelligent® Parts
  9. Do we fail to exert self-control because we lack resources or motivation? Competing theories to explain a debated phenomenon
  10. Representative time use data and new harmonised calibration of the American Heritage Time Use Data (AHTUD) 1965-1999
  11. Choice and quantity demand for improved and unimproved public water sources in rural areas
  12. Digitization and path disruption
  13. Analog, Digital, and the Cybernetic Illusion
  14. Analysis of Dynamic Response of a Two Degrees of Freedom (2-DOF) Ball Bearing Nonlinear Model
  15. B7-H1 restricts neuroantigen-specific T cell responses and confines inflammatory CNS damage: implications for the lesion pathogenesis of multiple sclerosis.
  16. Ensuring the Long-Term Provision of Heathland Ecosystem Services—The Importance of a Functional Perspective in Management Decision Frameworks
  17. Correction to
  18. Design rules for environmental biodegradability of phenylalanine alkyl ester linked ionic liquids
  19. Does symbolic representation through class signalling appeal to voters? Evidence from a conjoint experiment
  20. Relative wage positions and quit behavior
  21. Frame-based Optimal Design
  22. Microstructure evolution of Mg-11Gd-4.5Y-1Nd-1.5Zn-0.5Zr (wt%) alloy during deformation and its effect on strengthening
  23. The polarity field concept
  24. UAV remote laser scanner improvement by continuous scanning using DC motors
  25. A specification schema for software connectors