LC-Protonets: Multi-label Few-shot learning for world music audio tagging

Charilaos Papaioannou, Emmanouil Benetos, Alexandros Potamianos
{"title":"LC-Protonets: Multi-label Few-shot learning for world music audio tagging","authors":"Charilaos Papaioannou, Emmanouil Benetos, Alexandros Potamianos","doi":"arxiv-2409.11264","DOIUrl":null,"url":null,"abstract":"We introduce Label-Combination Prototypical Networks (LC-Protonets) to\naddress the problem of multi-label few-shot classification, where a model must\ngeneralize to new classes based on only a few available examples. Extending\nPrototypical Networks, LC-Protonets generate one prototype per label\ncombination, derived from the power set of labels present in the limited\ntraining items, rather than one prototype per label. Our method is applied to\nautomatic audio tagging across diverse music datasets, covering various\ncultures and including both modern and traditional music, and is evaluated\nagainst existing approaches in the literature. The results demonstrate a\nsignificant performance improvement in almost all domains and training setups\nwhen using LC-Protonets for multi-label classification. In addition to training\na few-shot learning model from scratch, we explore the use of a pre-trained\nmodel, obtained via supervised learning, to embed items in the feature space.\nFine-tuning improves the generalization ability of all methods, yet\nLC-Protonets achieve high-level performance even without fine-tuning, in\ncontrast to the comparative approaches. We finally analyze the scalability of\nthe proposed method, providing detailed quantitative metrics from our\nexperiments. The implementation and experimental setup are made publicly\navailable, offering a benchmark for future research.","PeriodicalId":501284,"journal":{"name":"arXiv - EE - Audio and Speech Processing","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - EE - Audio and Speech Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11264","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

We introduce Label-Combination Prototypical Networks (LC-Protonets) to address the problem of multi-label few-shot classification, where a model must generalize to new classes based on only a few available examples. Extending Prototypical Networks, LC-Protonets generate one prototype per label combination, derived from the power set of labels present in the limited training items, rather than one prototype per label. Our method is applied to automatic audio tagging across diverse music datasets, covering various cultures and including both modern and traditional music, and is evaluated against existing approaches in the literature. The results demonstrate a significant performance improvement in almost all domains and training setups when using LC-Protonets for multi-label classification. In addition to training a few-shot learning model from scratch, we explore the use of a pre-trained model, obtained via supervised learning, to embed items in the feature space. Fine-tuning improves the generalization ability of all methods, yet LC-Protonets achieve high-level performance even without fine-tuning, in contrast to the comparative approaches. We finally analyze the scalability of the proposed method, providing detailed quantitative metrics from our experiments. The implementation and experimental setup are made publicly available, offering a benchmark for future research.
LC-Protonets:用于世界音乐音频标记的多标签少镜头学习
我们引入了标签组合原型网络(LC-Protonets)来解决多标签少量分类问题,在这种情况下,模型必须根据少量可用示例归纳出新的类别。LC-Protonets 对原型网络进行了扩展,根据有限训练项目中存在的强大标签集,为每个标签组合生成一个原型,而不是为每个标签生成一个原型。我们的方法被应用于各种音乐数据集的自动音频标记,涵盖各种文化,包括现代音乐和传统音乐,并与文献中的现有方法进行了对比评估。结果表明,在使用 LC-Protonets 进行多标签分类时,几乎所有领域和训练设置的性能都有显著提高。微调提高了所有方法的泛化能力,但是 LC-Protonets 即使不进行微调也能获得高水平的性能,这与其他方法形成了鲜明对比。最后,我们分析了所提方法的可扩展性,并提供了实验中的详细量化指标。我们公开了实现方法和实验设置,为未来的研究提供了一个基准。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信