TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting

混合(物理) 系列(地层学) 时间序列 计量经济学 计算机科学 统计物理学 数学 地质学 物理 机器学习 古生物学 量子力学
作者
Shiyu Wang,Haixu Wu,Xiaoming Shi,Tengge Hu,Huakun Luo,Lintao Ma,James Y. Zhang,Jun Zhou
出处
期刊:Cornell University - arXiv 被引量:26
标识
DOI:10.48550/arxiv.2405.14616
摘要

Time series forecasting is widely used in extensive applications, such as traffic planning and weather forecasting. However, real-world time series usually present intricate temporal variations, making forecasting extremely challenging. Going beyond the mainstream paradigms of plain decomposition and multiperiodicity analysis, we analyze temporal variations in a novel view of multiscale-mixing, which is based on an intuitive but important observation that time series present distinct patterns in different sampling scales. The microscopic and the macroscopic information are reflected in fine and coarse scales respectively, and thereby complex variations can be inherently disentangled. Based on this observation, we propose TimeMixer as a fully MLP-based architecture with Past-Decomposable-Mixing (PDM) and Future-Multipredictor-Mixing (FMM) blocks to take full advantage of disentangled multiscale series in both past extraction and future prediction phases. Concretely, PDM applies the decomposition to multiscale series and further mixes the decomposed seasonal and trend components in fine-to-coarse and coarse-to-fine directions separately, which successively aggregates the microscopic seasonal and macroscopic trend information. FMM further ensembles multiple predictors to utilize complementary forecasting capabilities in multiscale observations. Consequently, TimeMixer is able to achieve consistent state-of-the-art performances in both long-term and short-term forecasting tasks with favorable run-time efficiency.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
刚刚
刚刚
SciGPT应助晚来天欲雪采纳,获得10
1秒前
cccyyy发布了新的文献求助10
1秒前
Sunsets完成签到 ,获得积分10
1秒前
2秒前
ZXB应助随梦而飞采纳,获得30
3秒前
脑洞疼应助CCC采纳,获得10
3秒前
梓歆发布了新的文献求助10
3秒前
lxzk11110000完成签到,获得积分10
4秒前
5秒前
powell发布了新的文献求助20
5秒前
wts完成签到,获得积分10
5秒前
CodeCraft应助liuxuying采纳,获得10
6秒前
卡卡发布了新的文献求助30
6秒前
8秒前
追寻易云发布了新的文献求助10
8秒前
9秒前
bkagyin应助MTF采纳,获得10
9秒前
背后书芹发布了新的文献求助10
11秒前
11秒前
maox1aoxin应助lxzk11110000采纳,获得30
12秒前
13秒前
没假期发布了新的文献求助10
14秒前
忍蛙发布了新的文献求助10
14秒前
lc完成签到,获得积分10
15秒前
kryptonite发布了新的文献求助10
15秒前
小龙仔123应助科研通管家采纳,获得10
15秒前
orixero应助科研通管家采纳,获得10
15秒前
15秒前
脑洞疼应助科研通管家采纳,获得10
16秒前
16秒前
16秒前
16秒前
16秒前
16秒前
16秒前
17秒前
17秒前
powell完成签到,获得积分20
18秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
Picture Books with Same-sex Parented Families: Unintentional Censorship 700
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
不知道标题是什么 500
Indomethacinのヒトにおける経皮吸収 400
Effective Learning and Mental Wellbeing 400
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3975953
求助须知:如何正确求助?哪些是违规求助? 3520269
关于积分的说明 11201866
捐赠科研通 3256738
什么是DOI,文献DOI怎么找? 1798436
邀请新用户注册赠送积分活动 877578
科研通“疑难数据库(出版商)”最低求助积分说明 806464