Byzantine-Robust Federated Fire Detection with a Rotating Coordinator
Not specified in the provided content
Abstract
The paper addresses the challenges of applying federated learning to indoor fire detection systems, focusing on issues like limited bandwidth, Byzantine clients, and reliance on a fixed aggregation server.
Reality Card
The proposed semi-decentralized Byzantine-robust federated learning method achieves comparable accuracy and detection speed to fixed-server methods while eliminating single points of failure.
Model updates are compressed up to 10 times with only a small loss in balanced accuracy.
The reproducibility of the results may be limited by the availability of the curated indoor fire-detection dataset and the complexity of the semi-decentralized method.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.