Revisiting OmniAnomaly for Anomaly Detection: performance metrics and comparison with PCA-based models

TL;DR

This paper critically evaluates OmniAnomaly for multivariate time series anomaly detection, revealing that simpler PCA-based models can perform comparably or better under consistent evaluation protocols, questioning the necessity of complex models.

Contribution

It provides a systematic comparison of OmniAnomaly and PCA-based models using standardized evaluation methods, highlighting the importance of methodology in assessing model performance.

Findings

PCA achieves comparable or better performance than OmniAnomaly in many cases.

Performance varies significantly across different machines and runs.

Evaluation methodology critically influences the perceived effectiveness of anomaly detection models.

Abstract

Deep learning models have become the dominant approach for multivariate time series anomaly detection (MTSAD), often reporting substantial performance improvements over classical statistical methods. However, these gains are frequently evaluated under heterogeneous thresholding strategies and evaluation protocols, making fair comparisons difficult. This work revisits OmniAnomaly, a widely used stochastic recurrent model for MTSAD, and systematically compares it with a simple linear baseline based on Principal Component Analysis (PCA) on the Server Machine Dataset (SMD). Both methods are evaluated under identical thresholding and evaluation procedures, with experiments repeated across 100 runs for each of the 28 machines in the dataset. Performance is evaluated using Precision, Recall and F1-score at point-level, with and without point-adjustment, and under different aggregation strategies across machines and runs, with the corresponding standard deviations also reported. The results show large variability across machines and show that PCA can achieve performance comparable to OmniAnomaly, and even outperform it when point-adjustment is not applied. These findings question the added value of more complex architectures under current benchmarking practices and highlight the critical role of evaluation methodology in MTSAD research.