Continual Learning For On-Device Environmental Sound Classification

TL;DR

This paper introduces an efficient continual learning method for on-device environmental sound classification that reduces computational costs by measuring classification uncertainty, enabling incremental learning without catastrophic forgetting.

Contribution

The proposed method uses uncertainty measurement based on classifier embedding perturbations to select historical data, improving efficiency and accuracy in continual learning for sound classification.

Findings

Outperforms baseline methods in accuracy

Reduces computational costs significantly

Effective in incremental learning scenarios

Abstract

Continuously learning new classes without catastrophic forgetting is a challenging problem for on-device environmental sound classification given the restrictions on computation resources (e.g., model size, running memory). To address this issue, we propose a simple and efficient continual learning method. Our method selects the historical data for the training by measuring the per-sample classification uncertainty. Specifically, we measure the uncertainty by observing how the classification probability of data fluctuates against the parallel perturbations added to the classifier embedding. In this way, the computation cost can be significantly reduced compared with adding perturbation to the raw data. Experimental results on the DCASE 2019 Task 1 and ESC-50 dataset show that our proposed method outperforms baseline continual learning methods on classification accuracy and computational efficiency, indicating our method can efficiently and incrementally learn new classes without the catastrophic forgetting problem for on-device environmental sound classification.