Interpretable Data Fusion for Distributed Learning: A Representative Approach via Gradient Matching

arXiv:2405.03782v1 Announce Type: new
Abstract: This paper introduces a representative-based approach for distributed learning that transforms multiple raw data points into a virtual representation. Unlike traditional distributed learning methods such as Federated Learning, which do not offer human interpretability, our method makes complex machine learning processes accessible and comprehensible. It achieves this by condensing extensive datasets into digestible formats, thus fostering intuitive human-machine interactions. Additionally, this approach maintains privacy and communication efficiency, and it matches the training performance of models …

abstract arxiv cs.hc cs.lg data distributed distributed learning federated learning fusion gradient human interpretability machine machine learning multiple paper processes raw raw data representation type via virtual