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.
First International Conference on the Design of Cyber-Secure Water Plants. 2024.

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Harvard

Paulsen, T & Brefeld, U 2024, Studying properties of water data using manifold-aware anomaly detectors. in First International Conference on the Design of Cyber-Secure Water Plants.

APA

Paulsen, T., & Brefeld, U. (2024). Studying properties of water data using manifold-aware anomaly detectors. In First International Conference on the Design of Cyber-Secure Water Plants

Vancouver

Paulsen T, Brefeld U. Studying properties of water data using manifold-aware anomaly detectors. in First International Conference on the Design of Cyber-Secure Water Plants. 2024

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. ",
author = "Tino Paulsen and Ulf Brefeld",
year = "2024",
month = apr,
day = "23",
language = "English",
booktitle = "First International Conference on the Design of Cyber-Secure Water Plants",

}

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.

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

BT - First International Conference on the Design of Cyber-Secure Water Plants

ER -