Multivariate long sequence time-series forecasting (M-LSTF) is a practical but challenging problem. Unlike traditional timer-series forecasting tasks, M-LSTF tasks are more challenging from two aspects: 1) M-LSTF models need to learn time-series patterns both within and between multiple time features; 2) Under the rolling forecasting setting, the similarity between two consecutive training samples increases with the increasing prediction length, which makes models more prone to overfitting. In this paper, we propose a generalizable memory-driven Transformer to target M-LSTF problems. Specifically, we first propose a global-level memory component to drive the forecasting procedure by integrating multiple time-series features. In addition, we adopt a progressive fashion to train our model to increase its generalizability, in which we gradually introduce Bernoulli noises to training samples. Extensive experiments have been performed on five different datasets across multiple fields. Experimental results demonstrate that our approach can be seamlessly plugged into varying Transformer-based models to improve their performances up to roughly 30%. Particularly, this is the first work to specifically focus on the M-LSTF tasks to the best of our knowledge.

翻译：多变量长序列时间序列预测（M-LSTF）是一个实际但具有挑战性的问题。与传统时间序列预测任务不同，M-LSTF任务的挑战性体现在两个方面：1）M-LSTF模型需要学习多个时间特征内部及相互之间的时间序列模式；2）在滚动预测设置下，相邻训练样本之间的相似度随着预测长度的增加而增大，这使得模型更容易过拟合。本文针对M-LSTF问题，提出了一种可泛化的记忆驱动Transformer。具体而言，我们首先设计了一个全局级记忆组件，通过整合多个时间序列特征来驱动预测过程。此外，我们采用渐进式训练方法提升模型的泛化能力，逐步向训练样本中引入伯努利噪声。我们在五个不同领域的数据集上进行了大量实验。实验结果表明，我们的方法可无缝嵌入多种基于Transformer的模型，并将其性能提升高达约30%。特别地，据我们所知，这是首项专门针对M-LSTF任务的研究工作。