From Similarity to Superiority: Channel Clustering for Time Series Forecasting

系列(地层学) 聚类分析 相似性(几何) 计算机科学 数据挖掘 时间序列 人工智能 频道(广播) 模式识别(心理学) 计量经济学 机器学习 数学 电信 地质学 图像(数学) 古生物学
作者
Jialin Chen,Jan Eric Lenssen,Aosong Feng,Weihua Hu,Matthias Fey,Leandros Tassiulas,Jure Leskovec,Rex Ying
出处
期刊:Cornell University - arXiv 被引量:5
标识
DOI:10.48550/arxiv.2404.01340
摘要

Time series forecasting has attracted significant attention in recent decades. Previous studies have demonstrated that the Channel-Independent (CI) strategy improves forecasting performance by treating different channels individually, while it leads to poor generalization on unseen instances and ignores potentially necessary interactions between channels. Conversely, the Channel-Dependent (CD) strategy mixes all channels with even irrelevant and indiscriminate information, which, however, results in oversmoothing issues and limits forecasting accuracy. There is a lack of channel strategy that effectively balances individual channel treatment for improved forecasting performance without overlooking essential interactions between channels. Motivated by our observation of a correlation between the time series model's performance boost against channel mixing and the intrinsic similarity on a pair of channels, we developed a novel and adaptable Channel Clustering Module (CCM). CCM dynamically groups channels characterized by intrinsic similarities and leverages cluster identity instead of channel identity, combining the best of CD and CI worlds. Extensive experiments on real-world datasets demonstrate that CCM can (1) boost the performance of CI and CD models by an average margin of 2.4% and 7.2% on long-term and short-term forecasting, respectively; (2) enable zero-shot forecasting with mainstream time series forecasting models; (3) uncover intrinsic time series patterns among channels and improve interpretability of complex time series models.
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
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
脑洞疼应助青云天采纳,获得10
1秒前
等一只ya发布了新的文献求助10
1秒前
2秒前
3秒前
Funny完成签到,获得积分10
6秒前
7秒前
9秒前
孙东玥发布了新的文献求助30
9秒前
彭于晏应助ZR采纳,获得10
9秒前
9秒前
9秒前
英姑应助dingby采纳,获得10
10秒前
11秒前
李翔完成签到,获得积分10
12秒前
12秒前
小禾雀完成签到,获得积分10
14秒前
科研通AI6.1应助江遇采纳,获得10
14秒前
小满完成签到,获得积分10
15秒前
Huilin0327完成签到,获得积分10
15秒前
15秒前
鱼肠发布了新的文献求助10
16秒前
aco发布了新的文献求助10
16秒前
19秒前
21秒前
21秒前
无花果应助哈哈采纳,获得10
22秒前
等一只ya完成签到,获得积分10
22秒前
22秒前
23秒前
23秒前
25秒前
顾矜应助有生之年采纳,获得10
25秒前
领导范儿应助阿桐慕采纳,获得10
25秒前
26秒前
泥豪泥嚎发布了新的文献求助10
27秒前
00gi发布了新的文献求助10
28秒前
Monik发布了新的文献求助10
29秒前
喝水的鱼关注了科研通微信公众号
31秒前
32秒前
33秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
Modern Epidemiology, Fourth Edition 5000
Digital Twins of Advanced Materials Processing 2000
Weaponeering, Fourth Edition – Two Volume SET 2000
Polymorphism and polytypism in crystals 1000
Social Cognition: Understanding People and Events 800
Signals, Systems, and Signal Processing 610
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 纳米技术 有机化学 物理 生物化学 化学工程 计算机科学 复合材料 内科学 催化作用 光电子学 物理化学 电极 冶金 遗传学 细胞生物学
热门帖子
关注 科研通微信公众号,转发送积分 6026642
求助须知:如何正确求助?哪些是违规求助? 7671072
关于积分的说明 16183503
捐赠科研通 5174596
什么是DOI,文献DOI怎么找? 2768824
邀请新用户注册赠送积分活动 1752199
关于科研通互助平台的介绍 1638071