A privacy-preserving, distributed and cooperative FCM-based learning approach for Cancer Research

The article presents a novel approach to Distributed Artificial Intelligence, using Federated Learning and Fuzzy Cognitive Maps (FCM) for cancer diagnosis. This method allows multiple participants to learn an FCM model while preserving data privacy, in compliance with data privacy regulations. The novelty lies in the use of a Particle Swarm Optimization-based FCM learning in a distributed way. The method is applied to a cancer diagnosis problem and the results show improved performance compared to non-distributed methods.

Publication date: 15 Feb 2024
Project Page: https://arxiv.org/abs/2402.10102v1
Paper: https://arxiv.org/pdf/2402.10102