I4CS 2023 Innovations for Community services

This week, Prof. Dmitrienko is giving a keynote talk at I4CS 2023 Innovations for Community services in Bamberg. 

Machine Learning (ML) methods have reached a level of maturity where they are being widely deployed across various domains, aiding users in classification and decision-making tasks. In this presentation, we will showcase the numerous advantages ML offers for applications dedicated to detecting security threats on mobile platforms. However, it is important to address the security and privacy concerns that arise when utilizing ML methods. One particular focus of our talk will be on Federated Learning (FL), which is a distributed form of ML that enhances privacy preservation during the training of ML models. We will conduct a comprehensive evaluation of the security and privacy risks associated with FL, delving into the intricacies of targeted and untargeted poisoning attacks, as well as the countermeasures employed to mitigate these threats. Our discussion will highlight the ongoing challenges in this field, such as the ability to differentiate between poisoned models and benign but uncommon models, particularly those trained on datasets with different data distributions. We will also address the issue of adaptive attackers who, once aware of the detection method, can add an additional training loss to minimize any changes in the detection metric, effectively evading detection. To stimulate further dialogue and exploration, we will outline promising research directions and open avenues for future research work.