# Improving the Classification of Rare Chords with Unlabeled Data

Source: [Improving the Classification of Rare Chords with Unlabeled Data](<https://www.sota2.com/research/paper/68fa69511e209131cea21812>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T22:57:51.173Z

Authors: Marcelo Bortolozzo, Rodrigo Schramm, Claudio R. Jung

Published: 2020-12-13T13:02:55.000Z

## Abstract

In this work, we explore techniques to improve performance for rare classes in the task of Automatic Chord Recognition (ACR). We first explored the use of the focal loss in the context of ACR, which was originally proposed to improve the classification of hard samples. In parallel, we adapted a self-learning technique originally designed for image recognition to the musical domain. Our experiments show that both approaches individually (and their combination) improve the recognition of rare chords, but using only self-learning with noise addition yields the best results.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Automatic Chord Estimation on ACR (test)](<https://www.sota2.com/research/sota/automatic-chord-estimation-on-acr-test.md>): Automatic Chord Estimation; ACR (test); Root Accuracy: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2012.07055>)

[PDF](<https://arxiv.org/pdf/2012.07055.pdf>)