TransNAS-TSAD: Harnessing Transformers for Multi-Objective Neural Architecture Search in Time Series Anomaly Detection

arXiv:2311.18061v3 Announce Type: replace
Abstract: The surge in real-time data collection across various industries has underscored the need for advanced anomaly detection in both univariate and multivariate time series data. This paper introduces TransNAS-TSAD, a framework that synergizes the transformer architecture with neural architecture search (NAS), enhanced through NSGA-II algorithm optimization. This approach effectively tackles the complexities of time series data, balancing computational efficiency with detection accuracy. Our evaluation reveals that TransNAS-TSAD surpasses conventional anomaly detection models due to its …

abstract advanced anomaly anomaly detection architecture arxiv collection cs.lg cs.ne data data collection detection framework industries multi-objective multivariate nas neural architecture search paper real-time search series time data time series transformer transformer architecture transformers type