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

TL;DR

This paper presents a novel interpretable data fusion method for distributed learning that creates virtual data representations, maintaining privacy and performance while enhancing human interpretability and interaction.

Contribution

It introduces a representative-based approach that transforms raw data into interpretable virtual representations, improving transparency and efficiency over traditional federated learning methods.

Findings

Achieves comparable or better accuracy than federated learning.

Maintains privacy and communication efficiency.

Effective in complex models with many clients.

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 using raw data. Simulation results show that our approach is competitive with or outperforms traditional Federated Learning in accuracy and convergence, especially in scenarios with complex models and a higher number of clients. This framework marks a step forward in integrating human intuition with machine intelligence, which potentially enhances human-machine learning interfaces and collaborative efforts.