arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2605.18657cs.LGcs.AI
收藏

KairosHope: 一种基于双记忆架构的下一代时间序列基础模型,用于专门分类

KairosHope: A Next-Generation Time-Series Foundation Model for Specialized Classification via Dual-Memory Architecture

Department of Computer Science and Artificial Intelligence(计算机科学与人工智能系);DiCITS, iMUDS, DaSCI(DiCITS、iMUDS、DaSCI);University of Granada(格拉纳达大学);Advanced Medical Imaging Group(先进医学成像组);Instituto de Investigación Biosanitaria de Granada (ibs.Granada)(格拉纳达生物医学研究机构(ibs.Granada));Department of Software Engineering(软件工程系);Department of Rural Engineering(农村工程系);University of Córdoba(科尔多瓦大学)

Luis Balderas, José Alberto Rodríguez, Miguel Lastra, Antonio Arauzo-Azofra, José M. Benítez

AI总结:

针对标准注意力计算瓶颈和经典统计知识缺失问题,提出KairosHope模型,通过双记忆系统(Titans模块和连续记忆系统CMS)替代二次注意力,并融合深度表示与统计特征的混合决策头,在UCR基准上实现优越分类性能。

摘要:

时间序列基础模型(TSFMs)在通用预测任务中取得了显著成功;然而,它们对专门分类问题的适应仍然受到标准注意力的计算瓶颈和对经典统计知识的系统性忽略的限制。本技术报告介绍了KairosHope,一种下一代TSFM,旨在协调大规模泛化与分类任务中的分析精度。该提案的核心是HOPE块,一种用双记忆系统替代二次注意力的架构:用于动态短期保留的Titans模块和用于长期历史上下文抽象的连续记忆系统(CMS)。为了丰富归纳偏差,引入了混合决策头,它将深度潜在表示与通过tsfeatures包提取的确定性统计特征融合。KairosHope在大型Monash档案上进行自监督预训练,结合了掩码时间序列建模(MTSM)和对比学习(InfoNCE)。随后,通过严格的线性探测和全微调(LP-FT)协议在UCR基准数据集上进行适应,以防止灾难性遗忘。实验结果表明,在具有严格时间因果关系的领域(如HAR或传感器数据)中,性能优越。因此,KairosHope为基础模型适应时间序列分析建立了一个稳健高效的框架。

英文摘要:

Time Series Foundation Models (TSFMs) have demonstrated notable success in general-purpose forecasting tasks; however, their adaptation to specialized classification problems remains constrained by the computational bottleneck of standard attention and the systematic omission of classical statistical knowledge. This technical report introduces KairosHope, a next-generation TSFM designed to reconcile massive generalization with analytical precision in classification tasks. The core of the proposal is the HOPE block, an architecture that replaces quadratic attention with a dual-memory system: Titans modules for dynamic short-term retention and a Continuum Memory System (CMS) for the abstraction of long-term historical context. To enrich the inductive bias, a Hybrid Decision Head is introduced, which fuses deep latent representations with deterministic statistical features extracted via tsfeatures package. KairosHope undergoes self-supervised pre-training on the massive Monash archive, combining Masked Time Series Modeling (MTSM) and contrastive learning (InfoNCE). Its subsequent adaptation to the UCR benchmark datasets is conducted through a rigorous Linear Probing and Full Fine-Tuning (LP-FT) protocol to prevent catastrophic forgetting. Empirical results demonstrate superior performance in domains characterized by strict temporal causality such as HAR or Sensor data. Consequently, KairosHope establishes a robust and efficient framework for the adaptation of foundation models to time series analysis.

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。

↑