The most popular Graph Neural Networks (GNNs) rely on the message-passing paradigm, where the idea is to iteratively propagate the representation information of every node to its direct neighborhood. However, in their most natural version, their expressive power is limited due to their connection to the Weisfeiler-Lehman test, a problem closely tied to the one of graph isomorphism. In this …
Dynamic Proactive Secret Sharing for Confidential Byzantine Fault-Tolerant Services using Multi-party Computation
New methods for blockchain technology and decentralized infrastructures emerged using Byzantine Fault-Tolerant (BFT) State Machine Replication (SMR), a reliable approach for creating trustworthy services. However, BFT SMR lacks secrecy protection, though it ensures integrity and availability even in the presence of live adversaries. Earlier attempts to address this issue integrated secret sharing into the consensus-based framework of BFT SMR but …
Research of the quality of water and sediments in the Silistra region of the Danube
The Danube River, spanning 19 countries, faces pollution risks from various sources. Its importance extends beyond fishing, serving as recreation and sustenance. Despite the seriousness of pollution, community awareness is lacking. Monitoring efforts near Silistra focus on water quality, ignoring sediment evaluation. Analysis revealed high cadmium levels. Lack of Bulgarian sediment standards led to reliance on American and European data. …