Quantum Safe Federated Learning with Gossip Based Communication

View at publisher Summarise with VIRAT
Authors: D. Dharani, Mansi Sharma, Sathiyavathi T, Vibha VS, Harrish R
Cited by: 0
References: 15

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This ability of edge computing allows us to develop critical autonomous infrastructures with often communicating peers which rely on processing data closer to its point of origin. The reported work will evaluate the effectiveness of security and privacy mechanisms developed for distributed systems with a center of attention on the edge devices that, as workers, process distributed datasets among multiple clients. Comparisons are made to modified Federated Machine Learning (FML) techniques, especially designed to be Gossip verifiable and Quantum Safe, with the outcomes of prior research of Distributed Machine Learning (DML) techniques. Specifically, the secret sharing principle is incorporated with the conventional federated averaging algorithm along with the lattice based Nth degree Truncated Polynomial Ring Units (NTRU), FrodoKEM and Kyber and cryptosystems. This adaptation is to make the system immune to quantum attacks. Thus, the Gossip data dissemination protocol, running atop the distributed network via custom Ethereum blockchain, authenticates and validates the learning model propagated among the peers in the network. An additional layer of integrity is offered by it.

Fields of Study & Subjects

Privacy-Preserving Technologies in Data Cryptography and Data Security Blockchain Technology Applications and Security

Publication details

Journal / source
International Conference on Intelligent Systems for Communication, IoT and Security (ICISCoIS)
Pages
1–11
Publication date
8 January 2026
Language
EN