This workshop has been led by Pavlo Mozharovskyi, Telecom Paris.
Anomaly detection (Chandola et al., 2009) is a branch of Machine Learning which aims at identifying observations that exhibit abnormal behavior. Be it measurement errors, disease development, severe weather, production quality default(s) (items) or failed equipment, financial frauds or crisis events, their on-time identification, isolation and explanation constitute an important task in almost any branch of industry and science.
During this workshop, one of these following topic has been discussed :
- the concept of data depth in both functional and multivariate settings,
- review most common notion of the depth function (halfspace (Tukey, 1975), projection (Zuo & Sefling,2000), zonoid (Mosler, 2002), spatial depth (Koltchinskii, 1997); integrated (Claeskens et al., 2014) and curve (Lafaye De Micheaux et al., 2020) functional depths, functional isolation forest Staerman et al. (2019)
- focus on a number of real-world applications ranging from simulated situations to hurricane tracks and brain imaging.
If you want further information on the workshop, don't hesitate to watch the replay and to check our Github.
Atelier réalisé le 24 septembre 2021.