arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.00465cs.SDcs.CVcs.LGeess.AS
收藏

音频分析的持续学习场景与策略特征化

Characterizing Continual Learning Scenarios and Strategies for Audio Analysis

Indian Institute of Technology Ropar(印度理工学院罗帕尔分校);University of Regensburg(雷根斯堡大学);Inception Institute of AI(Inception人工智能研究院);Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI)(穆罕默德·本·扎耶德人工智能大学);University of Ottawa(渥太华大学)

Ruchi Bhatt, Pratibha Kumari, Dwarikanath Mahapatra, Abdulmotaleb El Saddik, Mukesh Saini

AI总结:

本文针对音频分析中数据分布漂移问题,构建了全面的持续学习数据集,系统评估了多种CL与非CL方法,发现Replay在DCASE数据上表现最佳。

摘要:

音频分析在许多应用场景中非常有用。最先进的音频分析方法假设训练和部署时的数据分布是相同的。然而,由于现实生活中的各种挑战,数据可能会遇到分布漂移,或者在不久的将来会遇到新的类别。因此,一次性训练的模型可能表现不佳。持续学习(CL)方法旨在处理数据分布中的此类变化。已有一些尝试将CL方法用于音频分析。然而,缺乏系统的评估框架。在本文中,我们创建了一个全面的CL数据集,并针对基于音频的监控任务对CL方法进行了特征化。我们研究了以下CL和非CL方法:EWC、LwF、SI、GEM、A-GEM、GDumb、Replay、Naive、Cumulative和Joint训练。该研究对从事音频分析领域以开发自适应模型的研究人员和从业者非常有益。我们观察到,Replay在DCASE挑战数据中取得了比其他方法更好的结果。它在域增量场景中达到了70.12%的准确率,在类增量场景中达到了96.98%的准确率。

英文摘要:

Audio analysis is useful in many application scenarios. The state-of-the-art audio analysis approaches assume the data distribution at training and deployment time will be the same. However, due to various real-life challenges, the data may encounter drift in its distribution or can encounter new classes in the late future. Thus, a one-time trained model might not perform adequately. Continual learning (CL) approaches are devised to handle such changes in data distribution. There have been a few attempts to use CL approaches for audio analysis. Yet, there is a lack of a systematic evaluation framework. In this paper, we create a comprehensive CL dataset and characterize CL approaches for audio-based monitoring tasks. We have investigated the following CL and non-CL approaches: EWC, LwF, SI, GEM, A-GEM, GDumb, Replay, Naive, Cumulative, and Joint training. The study is very beneficial for researchers and practitioners working in the area of audio analysis for developing adaptive models. We observed that Replay achieved better results than other methods in the DCASE challenge data. It achieved an accuracy of 70.12% for the domain incremental scenario and an accuracy of 96.98% for the class incremental scenario.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑