Mass Volume curves and anomaly ranking. S. Clémençon, A. Thomas. Electronic Journal of Statistics (2018). pdf
Anomaly detection in extreme regions via empirical MV-sets on the sphere. A. Thomas, S. Clémençon, A. Gramfort, A. Sabourin. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (2017). pdf
Learning hyperparameters for unsupervised anomaly detection. A. Thomas, S. Clémençon, V. Feuillard, A. Gramfort. Anomaly Detection Workshop, ICML 2016. Co-winner of Google best paper award. pdf
Calibration of One-Class SVM for MV set estimation. A. Thomas, V. Feuillard and A. Gramfort. In 2015 IEEE International Conference on Data Science and Advanced Analytics, DSAA 2015. pdf
Talks
Rare Events, Extremes and Machine Learning Workshop, 2018. The Mass Volume curve, a performance metric for unsupervised anomaly detection. slides
February 2018 Center for Data Science meeting. Anomaly detection in scikit-learn. slides
CAP 2016. Learning hyperparameters for unsupervised anomaly detection.
