"Modified Partial Data Method for Mutual Synchronization of Tree Parity Machines"
Mykyta Aleksandrov · Scientific papers of Donetsk National Technical University Series Informatics Cybernetics and Computer Science · 2025
"This article proposes an approach for the partial use of synapse values for the mutual synchronization of neural networks. Tree parity machines are used as mutually synchronizing neural networks. Synchronization completion is confirmed by comparing hashes calculated based on the synapse values of the neural networks. Partial use of synapse values is proposed to reduce the amount of data transmitted over the network in either open or encrypted form. The study experimentally confirmed the feasibility of this approach and identified an acceptable amount of unused data. Improvements to this method were also proposed and experimentally tested."