Symposium on anomaly detection

Subject: Symposium on Anomaly Detection (common event LUSIS-DSAIDIS)

Date: February 29, 2024

Place: Amphi IV, Eiffel building, CentraleSupélec

Registration link (by 25th February) : https://framaforms.org/colloque-detection-danomalies-chaire-lusis-2902-1705915390

Abstract: The first symposium on anomaly detection organized within the framework of the LUSIS Chair (https://chaire-lusis.centralesupelec.fr/), focuses on research work and applications for identifying non-conforming behavior in data sets. Lectures, presentations and a poster forum will present methods such as descriptive statistics, as well as approaches based on machine learning, including neural networks, transformers and autoencoder algorithms. ractical case studies will be presented, focusing on the application of these techniques in a variety of fields, such as finance, healthcare, transport, industry, and for example fraud detection and predictive maintenance. During a round table discussion, experts in the field will discuss anomaly detection and the link between research and business. Participants will have the opportunity to deepen their knowledge and exchange ideas on the scientific and industrial challenges linked to anomaly detection.

This symposium is supported by the DATAIA’s Institute

9am-9.10am: Introduction by Fabrice Popineau

9.10am-9h50am: ” Anomaly detection using data depth: multivariate setting ” by Pavlo Mozharovskyi

9.50am-10.30am: ” The alliance between Industry and Research for Fraud Detection ” by Fabrice Daniel

Break

11.00am-12.00am: Oral presentations of PhD students’ work

12.00am-1.00pm: Round table with experts in the field + discussion with the audience

1.00pm-2.30pm: Poster session + cocktail

2:30pm: Conclusion