arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2307.09019cs.LGcs.CV
收藏

U-shaped Transformer:在时间序列分析中保留高频上下文

U-shaped Transformer: Retain High Frequency Context in Time Series Analysis

School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology(上海理工大学光电信息与计算机工程学院)

Qingkui Chen, Yiqin Zhang

AI总结:

针对时间序列预测中transformer高频信息丢失、性能不及MLP的问题,提出融合MLP优势、引入Unet跳层连接与多尺度特征提取的U-shaped Transformer,以低成本在多数据集上取得先进效果。

摘要:

时间序列预测在各类工业领域中发挥着关键作用。近年来,以transformer为主干的神经网络在计算机视觉、自然语言处理等诸多领域取得了显著成效。在时间序列分析领域,已有研究表明,即便是最简单的MLP(多层感知机)网络,在时间序列预测任务上的表现也优于先进的基于transformer的网络。但我们认为,这些发现恰恰说明时间序列存在低秩特性。本文中,我们考虑到transformer的低通特性,尝试融合MLP的优势。我们受Unet启发,在传统transformer主干中引入跳层连接,从而将输入的高频上下文保留至输出,该模型即U-shaped Transformer。我们引入了patch合并与拆分操作来提取不同尺度的特征,并使用更大规模的数据集以充分发挥transformer主干的性能。我们的实验表明,该模型以相对较低的成本,在多个数据集上达到了先进水平。

英文摘要:

Time series prediction plays a crucial role in various industrial fields. In recent years, neural networks with a transformer backbone have achieved remarkable success in many domains, including computer vision and NLP. In time series analysis domain, some studies have suggested that even the simplest MLP networks outperform advanced transformer-based networks on time series forecast tasks. However, we believe these findings indicate there to be low-rank properties in time series sequences. In this paper, we consider the low-pass characteristics of transformers and try to incorporate the advantages of MLP. We adopt skip-layer connections inspired by Unet into traditional transformer backbone, thus preserving high-frequency context from input to output, namely U-shaped Transformer. We introduce patch merge and split operation to extract features with different scales and use larger datasets to fully make use of the transformer backbone. Our experiments demonstrate that the model performs at an advanced level across multiple datasets with relatively low cost.

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。

↑