Temporal-Conditioned Normalizing Flows for Multivariate Time Series Anomaly Detection

TL;DR

This paper proposes temporal-conditioned normalizing flows (tcNF), a novel framework for multivariate time series anomaly detection that models temporal dependencies and uncertainty with high accuracy and robustness.

Contribution

The paper introduces tcNF, a new autoregressive normalizing flow model conditioned on past observations for improved anomaly detection in time series.

Findings

Demonstrates high accuracy on diverse datasets

Shows robustness compared to existing methods

Provides open-source implementation for reproducibility

Abstract

This paper introduces temporal-conditioned normalizing flows (tcNF), a novel framework that addresses anomaly detection in time series data with accurate modeling of temporal dependencies and uncertainty. By conditioning normalizing flows on previous observations, tcNF effectively captures complex temporal dynamics and generates accurate probability distributions of expected behavior. This autoregressive approach enables robust anomaly detection by identifying low-probability events within the learned distribution. We evaluate tcNF on diverse datasets, demonstrating good accuracy and robustness compared to existing methods. A comprehensive analysis of strengths and limitations and open-source code is provided to facilitate reproducibility and future research.