Music Instrument Classification Reprogrammed

TL;DR

This paper introduces a reprogramming technique that adapts pre-trained neural networks for music instrument classification, effectively overcoming data scarcity and achieving competitive or superior performance with fewer training resources.

Contribution

The paper presents a novel reprogramming approach that repurposes pre-trained models for music instrument classification, reducing the need for extensive annotated data.

Findings

Reprogrammed models perform on par or better than state-of-the-art.

Reprogramming requires fewer training parameters.

Technique shows promise for data-scarce tasks.

Abstract

The performance of approaches to Music Instrument Classification, a popular task in Music Information Retrieval, is often impacted and limited by the lack of availability of annotated data for training. We propose to address this issue with "reprogramming," a technique that utilizes pre-trained deep and complex neural networks originally targeting a different task by modifying and mapping both the input and output of the pre-trained model. We demonstrate that reprogramming can effectively leverage the power of the representation learned for a different task and that the resulting reprogrammed system can perform on par or even outperform state-of-the-art systems at a fraction of training parameters. Our results, therefore, indicate that reprogramming is a promising technique potentially applicable to other tasks impeded by data scarcity.