Title: Anomaly Detection in Time-Series

Abstract: Anomaly detection in time-series focuses on identifying data points or patterns that deviate significantly from the expected behavior in data. This process is critical in many fields, ranging from fraud detection in finance to monitoring system health in IT, cybersecurity  and predictive maintenance in manufacturing. In this presentation, we will talk about KoopAGRU, a new deep learning model designed to tackle this problem by combining Fast Fourier Transform (FFT), Deep Dynamic Mode Decomposition (DeepDMD), and Koopman theory. This work is done with collaboration of Issam Ait Yahia.