如有侵权或其他问题，欢迎留言联系更正或删除。出处：IJCAI 2024 ，链接：GitHub - jackyue1994/sub_adjacent_transformer贡献：1. 提出：基于 “次邻域” 及 “注意力贡献” 的注意力学习机制，以增强异常、正常的区分；2. 首次将 “线性注意力” 及 “可学习的映射函数” 引入TSAD。1. 基本思想Time points usually have stronger connections with their neighbors and fewer connections with distant points.This characteristic is more pronounced for anomalies [Xu et al., 2022]. → If werely solely on subadjacent neighborhoodsto reconstruct time points, the reconstruction errors of anomalies willbecom
Sub - Adjacent Transformer — 对AT的有趣改进
发布时间：2026/7/20 20:37:15
如有侵权或其他问题，欢迎留言联系更正或删除。出处：IJCAI 2024 ，链接：GitHub - jackyue1994/sub_adjacent_transformer贡献：1. 提出：基于 “次邻域” 及 “注意力贡献” 的注意力学习机制，以增强异常、正常的区分；2. 首次将 “线性注意力” 及 “可学习的映射函数” 引入TSAD。1. 基本思想Time points usually have stronger connections with their neighbors and fewer connections with distant points.This characteristic is more pronounced for anomalies [Xu et al., 2022]. → If werely solely on subadjacent neighborhoodsto reconstruct time points, the reconstruction errors of anomalies willbecom
