Similarity search improves model choice for time series anomaly detection
Retrieval Over Training: Similarity search-based Model Selection for Time Series Anomaly Detection
Machine LearningArtificial Intelligence
Summary
Detecting unusual events in time series data is important but no single method always works best. The authors created a way to pick the best detection method for a new time series by finding similar past series and seeing what worked well for them. This makes picking models more accurate and robust, especially for new or different types of data. Their approach also saves time by selecting fewer models to try while still matching or beating bigger combinations.
What this means in practice
- •For data engineers: Select the most effective anomaly detector for new time series data by comparing to similar previously seen series performance profiles.
- •For manufacturing monitoring teams: Improve detection of unusual patterns in machine sensor data using similarity-guided model selection to reduce false alarms and missed anomalies.
Authors
Christos Panourgias, Roberto Stanzione, Adrien Petralia, Themis Palpanas, Paul Boniol
Abstract
Anomaly detection is a fundamental task for time-series analytics, with important implications for the downstream performance of many applications. Despite the large number of anomaly detection methods proposed in the literature, recent benchmark studies have shown that no single detector performs best across highly heterogeneous time series. Therefore, a practical and scalable solution is to develop a model-selection method that, for a given time series, selects the anomaly detector most likely to perform well. Nevertheless, the model selection approaches proposed in the literature suffer from a significant drop in accuracy when applied in Out-of-Distribution (OOD) settings. In this paper, we tackle the aforementioned limitation and propose RAMSAD, a retrieval-based framework for model selection in time-series anomaly detection. Given a new time series, our method queries a knowledge base of previously observed instances, retrieves the top-k most similar series, and transfers detector recommendations from their performance profiles. The framework can operate with standard similarity measures as well as embedding-based representations. Our results show that RAMSAD outperforms existing baselines in In-Distribution (ID) settings and is the most robust selector in OOD, where, when selecting a small subset of detectors, it matches or exceeds ensembles that run the entire detector pool, at a fraction of their cost. Overall, we demonstrate that similarity-based retrieval constitutes a strong and efficient foundation for model selection in time-series anomaly detection.