FedMLAC: Mutual Learning Driven Heterogeneous Federated Audio Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bai, Jun, Rana, Rajib, Wu, Di, Qu, Youyang, Tao, Xiaohui, Zhang, Ji, Busso, Carlos, Palaiahnakote, Shivakumara
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918111243927552
author Bai, Jun
Rana, Rajib
Wu, Di
Qu, Youyang
Tao, Xiaohui
Zhang, Ji
Busso, Carlos
Palaiahnakote, Shivakumara
author_facet Bai, Jun
Rana, Rajib
Wu, Di
Qu, Youyang
Tao, Xiaohui
Zhang, Ji
Busso, Carlos
Palaiahnakote, Shivakumara
contents Federated Learning (FL) offers a privacy-preserving framework for training audio classification (AC) models across decentralized clients without sharing raw data. However, Federated Audio Classification (FedAC) faces three major challenges: data heterogeneity, model heterogeneity, and data poisoning, which degrade performance in real-world settings. While existing methods often address these issues separately, a unified and robust solution remains underexplored. We propose FedMLAC, a mutual learning-based FL framework that tackles all three challenges simultaneously. Each client maintains a personalized local AC model and a lightweight, globally shared Plug-in model. These models interact via bidirectional knowledge distillation, enabling global knowledge sharing while adapting to local data distributions, thus addressing both data and model heterogeneity. To counter data poisoning, we introduce a Layer-wise Pruning Aggregation (LPA) strategy that filters anomalous Plug-in updates based on parameter deviations during aggregation. Extensive experiments on four diverse audio classification benchmarks, including both speech and non-speech tasks, show that FedMLAC consistently outperforms state-of-the-art baselines in classification accuracy and robustness to noisy data.
format Preprint
id arxiv_https___arxiv_org_abs_2506_10207
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FedMLAC: Mutual Learning Driven Heterogeneous Federated Audio Classification
Bai, Jun
Rana, Rajib
Wu, Di
Qu, Youyang
Tao, Xiaohui
Zhang, Ji
Busso, Carlos
Palaiahnakote, Shivakumara
Sound
Distributed, Parallel, and Cluster Computing
Audio and Speech Processing
Federated Learning (FL) offers a privacy-preserving framework for training audio classification (AC) models across decentralized clients without sharing raw data. However, Federated Audio Classification (FedAC) faces three major challenges: data heterogeneity, model heterogeneity, and data poisoning, which degrade performance in real-world settings. While existing methods often address these issues separately, a unified and robust solution remains underexplored. We propose FedMLAC, a mutual learning-based FL framework that tackles all three challenges simultaneously. Each client maintains a personalized local AC model and a lightweight, globally shared Plug-in model. These models interact via bidirectional knowledge distillation, enabling global knowledge sharing while adapting to local data distributions, thus addressing both data and model heterogeneity. To counter data poisoning, we introduce a Layer-wise Pruning Aggregation (LPA) strategy that filters anomalous Plug-in updates based on parameter deviations during aggregation. Extensive experiments on four diverse audio classification benchmarks, including both speech and non-speech tasks, show that FedMLAC consistently outperforms state-of-the-art baselines in classification accuracy and robustness to noisy data.
title FedMLAC: Mutual Learning Driven Heterogeneous Federated Audio Classification
topic Sound
Distributed, Parallel, and Cluster Computing
Audio and Speech Processing
url https://arxiv.org/abs/2506.10207