MMiC: Mitigating Modality Incompleteness in Clustered Federated Learning

TL;DR

MMiC introduces a novel framework to address modality incompleteness in Clustered Multimodal Federated Learning, improving performance by strategic client selection and dynamic aggregation, validated through extensive experiments.

Contribution

The paper proposes MMiC, a new method that mitigates missing modalities in MFL by parameter replacement, client selection optimization, and dynamic global aggregation.

Findings

Outperforms existing architectures in global performance.

Enhances personalized learning on multimodal datasets.

Effectively handles missing modalities in federated settings.

Abstract

In the era of big data, data mining has become indispensable for uncovering hidden patterns and insights from vast and complex datasets. The integration of multimodal data sources further enhances its potential. Multimodal Federated Learning (MFL) is a distributed approach that enhances the efficiency and quality of multimodal learning, ensuring collaborative work and privacy protection. However, missing modalities pose a significant challenge in MFL, often due to data quality issues or privacy policies across the clients. In this work, we present MMiC, a framework for Mitigating Modality incompleteness in MFL within the Clusters. MMiC replaces partial parameters within client models inside clusters to mitigate the impact of missing modalities. Furthermore, it leverages the Banzhaf Power Index to optimize client selection under these conditions. Finally, MMiC employs an innovative approach to dynamically control global aggregation by utilizing Markovitz Portfolio Optimization. Extensive experiments demonstrate that MMiC consistently outperforms existing federated learning architectures in both global and personalized performance on multimodal datasets with missing modalities, confirming the effectiveness of our proposed solution. Our code is available at https://github.com/gotobcn8/MMiC.