[论文解读] Less Is More: Fast Multivariate Time Series Forecasting with Light Sampling-oriented MLP Structures

Tianping Zhang, Yizhuo Zhang+5 more|arXiv (Cornell University)|Jul 4, 2022

Time Series Analysis and Forecasting被引用 88

一句话总结

LightTS 引入一种轻量级的基于 MLP 的模型，使用连续采样和区间采样来捕捉多变量时间序列中的短期与长期模式，在远低于 FLOPS 的计算量下实现最先进或具有竞争力的结果，并具有稳定的鲁棒性。

ABSTRACT

Multivariate time series forecasting has seen widely ranging applications in various domains, including finance, traffic, energy, and healthcare. To capture the sophisticated temporal patterns, plenty of research studies designed complex neural network architectures based on many variants of RNNs, GNNs, and Transformers. However, complex models are often computationally expensive and thus face a severe challenge in training and inference efficiency when applied to large-scale real-world datasets. In this paper, we introduce LightTS, a light deep learning architecture merely based on simple MLP-based structures. The key idea of LightTS is to apply an MLP-based structure on top of two delicate down-sampling strategies, including interval sampling and continuous sampling, inspired by a crucial fact that down-sampling time series often preserves the majority of its information. We conduct extensive experiments on eight widely used benchmark datasets. Compared with the existing state-of-the-art methods, LightTS demonstrates better performance on five of them and comparable performance on the rest. Moreover, LightTS is highly efficient. It uses less than 5% FLOPS compared with previous SOTA methods on the largest benchmark dataset. In addition, LightTS is robust and has a much smaller variance in forecasting accuracy than previous SOTA methods in long sequence forecasting tasks.

研究动机与目标

• 激发研究：简单、轻量级模型是否能够与复杂架构在多元时间序列预测方面不相上下。
• 引入两种降采样策略，在减少输入长度的同时尽量保留信息。
• 开发一种基于 MLP 的架构，在降采样的子序列和变量之间进行信息交换。
• 展示 LightTS 在多个基准上的效率、鲁棒性和预测准确性。

提出的方法

• 应用连续采样和区间采样将每条时间序列分割为不重叠的子序列，同时不丢弃令牌。
• 使用带瓶颈设计的信息交换块（IEBlocks）来执行时序和通道投影，以进行特征提取和序列间交互。
• 将来自两个采样路径及跨变量的特征结合起来，通过第三个 IEBlock 形成最终预测。
• 以端到端的轻量级架构进行训练，能够高效处理长序列。

实验结果

研究问题

• RQ1 一个纯 MLP 基模型在轻量降采样下是否能在多元时间序列预测中达到或超过最先进方法？
• RQ2 连续采样和区间采样是否能在提高效率的同时有效捕捉短期与长期模式？
• RQ3 LightTS 是否对长输入序列具有鲁棒性并能在多样化数据集上扩展？

主要发现

• LightTS 在八个基准数据集上实现了最先进或具有竞争力的结果，在长序列预测方面超越了若干基线。
• LightTS 在最大基准数据集上所需 FLOPS 少于之前 SOTA 方法的 5% 以内。
• LightTS 在多种设置下的每个训练轮训练速度显著快于强基线。
• LightTS 在不同随机种子下的预测准确性方差较低，表明在长序列任务中鲁棒性提升。