Presentation of results in collaborative intrusion detection at CUME's national thematic day

Our PI, Gregory, has been invited to present some of the results obtained during Sara’s Ph.D thesis at the National Day of CUME, a French association of IT staff across the whole higher education sector, which topic this year is Cybersecurity in Higher Education and Research. The program can be found on their website.

Gregory’s talk closed a long day of talks and discussions mainly focusing on cybersecurity governance and operations in the higher education sector, involving speakers from ANSSI, the Ministry of Higher Education and Research, IT staff across both universities, engineering schools and research agencies across France. Gregory presented the motivation for ML-based collaboration in detecting intrusions, including ongoing attack campaigns, before new victims arise. Leveraging the Federated Prototype Learning-based approach developed by Sara, Gregory explained how a network or security operator may monitor the network in a distributed and privacy-preserving manner across its clients, and provide an intrusion detection model that is able to consolidate heterogeneous attack knowledge provided by its clients. Although a research and prospective talk, the presentation was well-received by the audience.