Studying properties of water data using manifold-aware anomaly detectors

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Authors

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.
OriginalspracheEnglisch
TitelDesign of Cyber-Secure Water Plants
HerausgeberAditya Mathur, Jianying Zhou, Gauthama Raman
VerlagMDPI AG
PublikationsstatusEingereicht - 23.04.2024
VeranstaltungFirst International Conference on the design of cyber-secure water plants - DCS-Water'24 - The Water Tower, Buford, USA / Vereinigte Staaten
Dauer: 23.04.202424.04.2024
https://itrust.sutd.edu.sg/first-international-conference-on-the-design-of-cyber-secure-water-plants-dcs-water24/paper-submission-dcs-water24/

Zuletzt angesehen

Publikationen

  1. Participation in multi-level policy implementation: exploring the influence of governance culture
  2. Rapid Prototyping of a Mechatronic Engine Valve Controller for IC Engines
  3. Re-visiting Effectuation
  4. Improving mechanical properties of chip-based aluminum extrudates by integrated extrusion and equal channel angular pressing (iECAP)
  5. The frame of the game
  6. Failing and the perception of failure in student-driven transdisciplinary projects
  7. Passive Rotation of Rotational Joints and Its Computation Method
  8. Fermentative utilization of coffee mucilage using Bacillus coagulans and investigation of down-stream processing of fermentation broth for optically pure L(+)-lactic acid production
  9. Big Data - Characterizing an Emerging Research Field using Topic Models
  10. Improving Human-Machine Interaction
  11. The Making of MEZ - Multilingual Development:
  12. Governing Objects from a Distance
  13. Individual differences and cognitive load theory
  14. Applying Quarter-Vehicle Model Simulation for Road Elevation Measurements Utilizing the Vehicle Level Sensor
  15. Optimal control strategies for PMSM with a decoupling super twisting SMC and inductance estimation in the presence of saturation
  16. Tree mixtures mediate negative effects of introduced tree species on bird taxonomic and functional diversity
  17. CHANGING RECREATIONAL ACTIVITIES FOR REDUCING INSOMNIA SEVERITY? RESULTS FROM A SERIAL MEDIATION ANALYSIS ON THE IMPACT OF RECREATIONAL BEHAVIOR AS A MECHANISM OF CHANGE IN DIGITAL INTERVENTIONS FOR INSOMNIA
  18. A Two-Stage Sliding-Mode High-Gain Observer to Reduce Uncertainties and Disturbances Effects for Sensorless Control in Automotive Applications
  19. Action Errors, Error Management, and Learning in Organizations
  20. What is normal?
  21. A Stacked Planar Sensor Concept for Minimally Invasive Plasma Monitoring
  22. Masked Autoencoder Pretraining for Event Classification in Elite Soccer
  23. How to attract visitors with strategic, value-based experience design
  24. Assessing authenticity in modelling test items: deriving a theoretical model
  25. From Planning to Implementation: Top-Down and Bottom-Up Approaches for Collaborative Watershed Management
  26. Endemic predators, invasive prey and native diversity
  27. Properties of some overlapping self-similar and some self-affine measures