RAMSeS: Robust and Adaptive Model Selection for Time-Series Anomaly Detection Algorithms

2026-02-25

AI Summary

Time series anomaly detection remains challenging due to its strong contextual dependencies, which hinder the development of universally effective methods and cause single models to exhibit unstable performance across diverse domains. To address this issue, this work proposes RAMSeS, a novel framework that uniquely integrates a stacked ensemble optimized via genetic algorithms with an adaptive single-model selection mechanism. The latter leverages Thompson sampling, GAN-based robustness testing, and Monte Carlo simulation to balance the strengths of ensemble learning with domain-specific adaptability. Experimental results demonstrate that RAMSeS significantly outperforms existing approaches in terms of F1 score, exhibiting superior robustness and cross-domain generalization capabilities.

Technology Category

• Machine Learning: Ensemble Methods
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Time-series data vary widely across domains, making a universal anomaly detector impractical. Methods that perform well on one dataset often fail to transfer because what counts as an anomaly is context dependent. The key challenge is to design a method that performs well in specific contexts while remaining adaptable across domains with varying data complexities. We present the Robust and Adaptive Model Selection for Time-Series Anomaly Detection RAMSeS framework. RAMSeS comprises two branches: (i) a stacking ensemble optimized with a genetic algorithm to leverage complementary detectors. (ii) An adaptive model-selection branch identifies the best single detector using techniques including Thompson sampling, robustness testing with generative adversarial networks, and Monte Carlo simulations. This dual strategy exploits the collective strength of multiple models and adapts to dataset-specific characteristics. We evaluate RAMSeS and show that it outperforms prior methods on F1.

Problem

• time-series anomaly detection
• model selection
• domain adaptation
• context-dependent anomalies
• data heterogeneity

Innovation

• adaptive model selection
• stacking ensemble
• genetic algorithm
• Thompson sampling
• generative adversarial networks