Adaptive Multi-Scale Decomposition Framework for Time Series Forecasting

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hu, Yifan, Liu, Peiyuan, Zhu, Peng, Cheng, Dawei, Dai, Tao
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916689768087552
author Hu, Yifan
Liu, Peiyuan
Zhu, Peng
Cheng, Dawei
Dai, Tao
author_facet Hu, Yifan
Liu, Peiyuan
Zhu, Peng
Cheng, Dawei
Dai, Tao
contents Transformer-based and MLP-based methods have emerged as leading approaches in time series forecasting (TSF). While Transformer-based methods excel in capturing long-range dependencies, they suffer from high computational complexities and tend to overfit. Conversely, MLP-based methods offer computational efficiency and adeptness in modeling temporal dynamics, but they struggle with capturing complex temporal patterns effectively. To address these challenges, we propose a novel MLP-based Adaptive Multi-Scale Decomposition (AMD) framework for TSF. Our framework decomposes time series into distinct temporal patterns at multiple scales, leveraging the Multi-Scale Decomposable Mixing (MDM) block to dissect and aggregate these patterns in a residual manner. Complemented by the Dual Dependency Interaction (DDI) block and the Adaptive Multi-predictor Synthesis (AMS) block, our approach effectively models both temporal and channel dependencies and utilizes autocorrelation to refine multi-scale data integration. Comprehensive experiments demonstrate that our AMD framework not only overcomes the limitations of existing methods but also consistently achieves state-of-the-art performance in both long-term and short-term forecasting tasks across various datasets, showcasing superior efficiency. Code is available at https://github.com/TROUBADOUR000/AMD
format Preprint
id arxiv_https___arxiv_org_abs_2406_03751
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Adaptive Multi-Scale Decomposition Framework for Time Series Forecasting
Hu, Yifan
Liu, Peiyuan
Zhu, Peng
Cheng, Dawei
Dai, Tao
Machine Learning
Transformer-based and MLP-based methods have emerged as leading approaches in time series forecasting (TSF). While Transformer-based methods excel in capturing long-range dependencies, they suffer from high computational complexities and tend to overfit. Conversely, MLP-based methods offer computational efficiency and adeptness in modeling temporal dynamics, but they struggle with capturing complex temporal patterns effectively. To address these challenges, we propose a novel MLP-based Adaptive Multi-Scale Decomposition (AMD) framework for TSF. Our framework decomposes time series into distinct temporal patterns at multiple scales, leveraging the Multi-Scale Decomposable Mixing (MDM) block to dissect and aggregate these patterns in a residual manner. Complemented by the Dual Dependency Interaction (DDI) block and the Adaptive Multi-predictor Synthesis (AMS) block, our approach effectively models both temporal and channel dependencies and utilizes autocorrelation to refine multi-scale data integration. Comprehensive experiments demonstrate that our AMD framework not only overcomes the limitations of existing methods but also consistently achieves state-of-the-art performance in both long-term and short-term forecasting tasks across various datasets, showcasing superior efficiency. Code is available at https://github.com/TROUBADOUR000/AMD
title Adaptive Multi-Scale Decomposition Framework for Time Series Forecasting
topic Machine Learning
url https://arxiv.org/abs/2406.03751