Beyond Marginals: Learning Joint Spatio-Temporal Patterns for Multivariate Anomaly Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Roy, Padmaksha, Boker, Almuatazbellah, Mili, Lamine
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908545848442880
author Roy, Padmaksha
Boker, Almuatazbellah
Mili, Lamine
author_facet Roy, Padmaksha
Boker, Almuatazbellah
Mili, Lamine
contents In this paper, we aim to improve multivariate anomaly detection (AD) by modeling the \textit{time-varying non-linear spatio-temporal correlations} found in multivariate time series data . In multivariate time series data, an anomaly may be indicated by the simultaneous deviation of interrelated time series from their expected collective behavior, even when no individual time series exhibits a clearly abnormal pattern on its own. In many existing approaches, time series variables are assumed to be (conditionally) independent, which oversimplifies real-world interactions. Our approach addresses this by modeling joint dependencies in the latent space and decoupling the modeling of \textit{marginal distributions, temporal dynamics, and inter-variable dependencies}. We use a transformer encoder to capture temporal patterns, and to model spatial (inter-variable) dependencies, we fit a multi-variate likelihood and a copula. The temporal and the spatial components are trained jointly in a latent space using a self-supervised contrastive learning objective to learn meaningful feature representations to separate normal and anomaly samples.
format Preprint
id arxiv_https___arxiv_org_abs_2509_15033
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Beyond Marginals: Learning Joint Spatio-Temporal Patterns for Multivariate Anomaly Detection
Roy, Padmaksha
Boker, Almuatazbellah
Mili, Lamine
Machine Learning
In this paper, we aim to improve multivariate anomaly detection (AD) by modeling the \textit{time-varying non-linear spatio-temporal correlations} found in multivariate time series data . In multivariate time series data, an anomaly may be indicated by the simultaneous deviation of interrelated time series from their expected collective behavior, even when no individual time series exhibits a clearly abnormal pattern on its own. In many existing approaches, time series variables are assumed to be (conditionally) independent, which oversimplifies real-world interactions. Our approach addresses this by modeling joint dependencies in the latent space and decoupling the modeling of \textit{marginal distributions, temporal dynamics, and inter-variable dependencies}. We use a transformer encoder to capture temporal patterns, and to model spatial (inter-variable) dependencies, we fit a multi-variate likelihood and a copula. The temporal and the spatial components are trained jointly in a latent space using a self-supervised contrastive learning objective to learn meaningful feature representations to separate normal and anomaly samples.
title Beyond Marginals: Learning Joint Spatio-Temporal Patterns for Multivariate Anomaly Detection
topic Machine Learning
url https://arxiv.org/abs/2509.15033