通过灵活GMRES稳定随机GMRES
音频与语音处理
2025-06-24 v1
摘要
我们探讨了将sketched GMRES作为外层包装器用于随机GMRES。基于对FGMRES残差相对于预条件器残差的新型不等式，我们推导出一种需要极少参数调节、仍具高效和鲁棒性（即生成非递增残差范数）的实用随机求解器。
引用
@article{arxiv.2506.18406,
title = {Fully Few-shot Class-incremental Audio Classification Using Multi-level Embedding Extractor and Ridge Regression Classifier},
author = {Yongjie Si and Yanxiong Li and Jiaxin Tan and Qianhua He and Il-Youp Kwak},
journal= {arXiv preprint arXiv:2506.18406},
year = {2025}
}
备注
Accepted for publication on Interspeech 2025. 5 pages, 6 tables, 7 figures