After a peer review process, a scientific paper from UBITECH’s Digital Security and Trusted Computing research group has been accepted for publication at the 19th Annual International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT 2023) that takes place in Cyprus, at the Coral Bay near the city of Pafos. UBITECH’s Dr Dimitrios Papamartzivanos, Entso Veliou, Dr Thanassis Giannetsos (Head of Group) and their co-authors suggest that the usage of explainable artificial intelligence (XAI) can lead to even more efficient performance as they tackle the limitation of conventional black box models, by enabling the understanding and interpretation of a model’s operation.
In particular, they address the plausible relation between the sensory input from devices over time via the application and integration of explainable artificial intelligence (XAI). XAI is exploited for the purpose of gaining knowledge and insights regarding the temporal data set that typically impose restrictions on human comprehension. Thus, their approach enables the reasoning of the model’s accuracy in the presence of adversaries and has the ability to shun out faulty or malicious data, thus, enhancing the model’s adaptation process. To this end, they provide a prototype implementation coupled with a detailed performance evaluation under different scenarios of attacks, employing both real and synthetic datasets. Our results suggest that the use of XAI leads to improved performance compared to other existing schemes.
