用频域神经网络复现并改进了时序预测模型FITS，提升对周期性模式的捕捉能力。
Self-Supervised Learning for Time Series: A Review & Critique of FITS
- 在复数频域训练单层神经网络，简化结构提升效率
- 在多变量回归任务中超越现有开源模型表现，尤其擅长周期性数据
- 提出两种融合DLinear的新方法，改善原模型对趋势和随机行为的建模缺陷
准确的时序预测在众多产业中具有重要价值。尽管深度学习不断进步，模型复杂度与规模持续上升，但许多先进模型的表现却仅与简单模型相当甚至更差。以近期提出的FITS为例，其宣称以极低参数量实现竞争性性能。我们通过在复数频域训练单层神经网络，成功复现了该结果。在多个真实世界数据集上的实验表明，FITS特别擅长捕捉周期性和季节性模式，但在趋势、非周期或类随机行为上表现不佳。为此，我们提出两种新型混合方法，将FITS与DLinear结合，不仅显著优于FITS独立使用，在多变量回归任务中成为当前最佳开源模型，且在价格数据集的多/线性回归任务中也取得优异成果。
原文摘要 · Abstract (English)
Accurate time series forecasting is a highly valuable endeavour with applications across many industries. Despite recent deep learning advancements, increased model complexity, and larger model sizes, many state-of-the-art models often perform worse or on par with simpler models. One of those cases is a recently proposed model, FITS, claiming competitive performance with significantly reduced parameter counts. By training a one-layer neural network in the complex frequency domain, we are able to replicate these results. Our experiments on a wide range of real-world datasets further reveal that FITS especially excels at capturing periodic and seasonal patterns, but struggles with trending, non-periodic, or random-resembling behavior. With our two novel hybrid approaches, where we attempt to remedy the weaknesses of FITS by combining it with DLinear, we achieve the best results of any known open-source model on multivariate regression and promising results in multiple/linear regression on price datasets, on top of vastly improving upon what FITS achieves as a standalone model.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。