Louise Travé-Massuyès

CLOE: Christoffel Loss Autoencoder for Anomaly Detection

Léa Billet, Louise Travé-Massuyès, Elodie Chanthery, Alexandre Gaffet

Semi-supervised anomaly detection plays a key role in diverse fields such as process monitoring, healthcare, and finance. However, lightweight methods often struggle with high-dimensional data and typically require careful tuning of multiple hyperparameters. Among existing approa…Read more