MUDAS: Mote-scale Unsupervised Domain Adaptation in Multi-label Sound Classification

TL;DR

MUDAS is a novel unsupervised domain adaptation framework designed for multi-label urban sound classification on resource-limited IoT devices, improving accuracy with minimal computational overhead.

Contribution

It introduces a resource-efficient UDA method that employs selective retraining, class-specific thresholds, and diversity regularization for multi-label sound classification in IoT environments.

Findings

Significant accuracy improvements over existing UDA methods.

Effective adaptation in resource-constrained IoT devices.

Robust performance across diverse urban sound datasets.

Abstract

Unsupervised Domain Adaptation (UDA) is essential for adapting machine learning models to new, unlabeled environments where data distribution shifts can degrade performance. Existing UDA algorithms are designed for single-label tasks and rely on significant computational resources, limiting their use in multi-label scenarios and in resource-constrained IoT devices. Overcoming these limitations is particularly challenging in contexts such as urban sound classification, where overlapping sounds and varying acoustics require robust, adaptive multi-label capabilities on low-power, on-device systems. To address these limitations, we introduce Mote-scale Unsupervised Domain Adaptation for Sounds (MUDAS), a UDA framework developed for multi-label sound classification in resource-constrained IoT settings. MUDAS efficiently adapts models by selectively retraining the classifier in situ using high-confidence data, minimizing computational and memory requirements to suit on-device deployment. Additionally, MUDAS incorporates class-specific adaptive thresholds to generate reliable pseudo-labels and applies diversity regularization to improve multi-label classification accuracy. In evaluations on the SONYC Urban Sound Tagging (SONYC-UST) dataset recorded at various New York City locations, MUDAS demonstrates notable improvements in classification accuracy over existing UDA algorithms, achieving good performance in a resource-constrained IoT setting.