The study, conducted by Jiaxi Yang et al., focuses on the challenges faced in querying multi-dimensional metadata on blockchain for selecting participants in Federated Learning (FL). The authors propose a novel data structure named MerkleRB-Tree to improve the query efficiency within each block. The method leverages Minimal Bounding Rectangle (MBR) and bloom-filters for the query process of multi-dimensional continuous-valued attributes and discrete-valued attributes respectively. The performance analysis and evaluation results demonstrate the superiority of this method in blockchain-based FL.
Publication date: 27 Sep 2023
Project Page: https://arxiv.org/abs/2309.15348
Paper: https://arxiv.org/pdf/2309.15348
