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 2607.02254eess.AS
收藏

跨域小样本增量音频分类:基于对抗对比学习

Cross Domain Few-Shot Class-Incremental Audio Classification Via Adversarial Contrastive Learning

School of Electronic and Information Engineering, South China University of Technology(华南理工大学电子与信息工程学院)

Yongjie Si, Yanxiong Li, Sen Huang, Beibei Liu

AI总结:

针对基类与增量类样本存在域偏移的跨域小样本增量音频分类问题,提出对抗对比训练策略,编码器在基会话训练后冻结,分类器持续更新,在六组跨域数据集上平均准确率超越现有方法。

摘要:

当前的小样本增量音频分类(FCAC)方法假设基类和增量类样本来自同一域(遵循相同分布)。然而,这两类样本之间通常存在域偏移。本文探索了跨域FCAC问题,其中基类和增量类样本存在域偏移。我们提出了一种对抗对比训练策略,使模型能够有效分类来自未见域的不同类别的样本。该模型由一个编码器和一个分类器组成。编码器在基会话中训练,但在增量会话中冻结,而分类器在所有会话中训练。在六对跨域数据集上进行了实验。结果表明,我们的方法在平均准确率上超过了最先进的方法。代码见此https URL。

英文摘要:

Current Few-shot Class-incremental Audio Classification (FCAC) methods assume that samples of base and incremental classes are in the same domain (following the same distribution). However, there is generally a domain shift between the above two types of samples. In this paper, we explore the problem of Cross Domain FCAC where samples of base and incremental classes have domain shift. We propose a strategy of adversarial contrastive training which enables the model to effectively classify samples of different classes from unseen domains. The model consists of an encoder and a classifier. The encoder is trained in base session but frozen in incremental sessions, whereas the classifier is trained in all sessions. Experiments are done on six pairs of cross-domain datasets. Results show that our method exceeds state-of-the-art methods in average accuracy. The code is at https://github.com/YongjieSi/ACL.

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。

↑