arXiv：cs.AI· Ziqiang Li, Yun Liu, Gouhei Tanaka·· 3 小时前
DTW-GBC：基于动态时间规整的粒度球计算实现鲁棒高效噪声标签时间序列分类
Robust and Efficient Noisy-Label Time-Series Classification via Dynamic Time Warping Based Granular Ball Computing
AI 导读
研究者提出基于动态时间规整的粒度球计算（DTW-GBC），将时序相似的训练样本组织成粒度球并在粒度层面完成分类，并开发了两种粒度球构建策略。在四个基准数据集的对称标签噪声实验中，两种 DTW-GBC 变体整体上缓解了标签噪声导致的性能下降，同时推理阶段所需的比较次数远少于基于 DTW 的 1-NN。
正文
Abstract:Dynamic Time Warping (DTW)-based Nearest-Neighbor (NN) classifiers are effective for time-series classification but are vulnerable to mislabeled training samples and require numerous DTW computations during inference. We propose DTW-based Granular Ball Computing (DTW-GBC), which organizes temporally similar training samples into granular balls and performs classification at the granule level. We further develop two granular-ball construction strategies for DTW-GBC. Experiments on four benchmark datasets with symmetric label noise show that the two DTW-GBC variants generally mitigate the performance degradation caused by label noise while requiring substantially fewer comparisons than DTW-based 1-NN during inference. These findings suggest that DTW-GBC provides a favorable balance between classification robustness and inference efficiency.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2608.11704 [cs.LG] |
| (or arXiv:2608.11704v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2608.11704 arXiv-issued DOI via DataCite |
Submission history
From: Ziqiang Li [view email]
[v1]
Wed, 12 Aug 2026 06:29:02 UTC (224 KB)
[v2]
Thu, 8 Oct 2026 08:12:17 UTC (224 KB)
来源：arXiv：cs.AI · arxiv.org