arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.35128cs.LG
收藏

STeP-TS:用于时间序列的带剪枝的共享时间滤波器

STeP-TS: Shared Temporal Filters with Pruning for Time Series

Albert-Ludwigs-Universität Freiburg(阿尔伯特-路德维希-弗莱堡大学)

Hanne Raum, Jasper Hoffmann, Joschka Boedecker

AI总结:

提出STeP-TS紧凑卷积时间序列预测框架,通过共享多尺度时间滤波器、变量权重及结构化剪枝,在性能相当的情况下减少参数与计算量。

摘要:

多变量时间序列预测需要捕捉跨多个尺度的时间依赖关系,同时兼顾共享动态与变量特定动态。紧凑的预测模型已成为日益庞大架构的有吸引力替代方案,但在严格的计算和内存预算内保持强预测性能仍具挑战性。我们提出一种紧凑的单层卷积框架,该框架为每个变量学习一组共享的多尺度时间滤波器。轻量级变量特定标量权重决定每个时间分量的贡献,使模型能够捕捉共享时间结构的同时适应变量特定动态。学习到的权重和滤波器输出能量进一步为卷积滤波器的结构化剪枝提供自然信号,在预测性能几乎不下降的情况下降低计算复杂度。我们还引入一种振幅自适应归一化机制,以保留变量依赖的时间振幅。在多个数据集上的实验表明,该模型的预测性能与最先进方法相当,同时所需参数和推理FLOPs显著更少,且实现了低样本延迟。

英文摘要:

Multivariate time series forecasting requires capturing temporal dependencies across multiple scales while accounting for both shared and variable-specific dynamics. Compact forecasting models have emerged as an attractive alternative to increasingly large architectures, but preserving strong predictive performance within tight computational and memory budgets remains challenging. We propose a compact one-layer convolutional framework that learns a shared bank of multi-scale temporal filters of each variable. Lightweight variable-specific scalar weights determine the contribution of each temporal component, enabling the model to capture shared temporal structure while adapting to variable-specific dynamics. The learned weights and filter-output energy further provide a natural signal for structured pruning of convolutional filters, reducing computational complexity with minimal degradation in forecasting performance. We additionally introduce an amplitude-adaptive normalization mechanism that preserves variable-dependent temporal amplitudes. Experiments on multiple datasets demonstrate forecasting performance competitive with state-of-the-art methods while requiring substantially fewer parameters and inference FLOPs and achieving low per-sample latency.

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。

↑