arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.03068cs.LGcs.AI
收藏

多域集成分布下的领域融合可控泛化用于跨领域时间序列预测

Domain Fusion Controllable Generalization for Cross-Domain Time Series Forecasting from Multi-Domain Integrated Distribution

State Key Laboratory for Novel Software Technology, Nanjing University(新型软件技术国家重点实验室,南京大学)

Xiangkai Ma, Xiaobin Hong, Mingkai Lin, Han Zhang, Wenzhong Li, Sanglu Lu

AI总结:

本文提出TimeControl模型,通过领域融合范式整合多领域时间序列信息,利用扩散模型实现跨领域时间序列预测的可控泛化。

摘要:

传统深度模型在时间序列预测中取得了前所未有的成功。然而,面对跨领域泛化挑战,现有研究利用统计先验作为提示工程,但在各领域之间巨大的分布偏移下失效。本文提出了一种新型时间序列泛化扩散模型(TimeControl),开创了领域融合范式,通过扩散模型系统地将多个时间序列领域的信息整合到统一的生成过程中。与捕捉预测时间跨度条件概率的自回归模型不同,我们利用扩散去噪过程来建模跨领域数据的混合分布,并通过条件采样直接生成目标领域的预测序列。所提出的TimeControl包含三个关键设计:(1)条件网络捕捉观测序列中的多尺度波动模式,这些模式被用作上下文表示以指导去噪网络生成预测序列;(2)基于适配器的微调策略,预训练阶段学习的多领域通用表示被用于目标领域的下游任务;(3)设计了一种新的混合架构以对齐观测和预测空间,使TimeControl能够灵活地生成任意长度的预测序列。我们在主流的49个基准和30个基线模型上进行了广泛的实验,结果表明TimeControl在所有数据领域均优于现有基线,展现出更强的零样本泛化能力。

英文摘要:

Conventional deep models have achieved unprecedented success in time series forecasting. However, facing the challenge of cross-domain generalization, existing studies utilize statistical prior as prompt engineering fails under the huge distribution shift among various domains. In this paper, a novel time series generalization diffusion model (TimeControl) that pioneers the Domain-Fusion paradigm, systematically integrating information from multiple time series domains into a unified generative process via diffusion models. Unlike the autoregressive models that capture the conditional probabilities of the prediction horizon to the historical sequence, we use the diffusion denoising process to model the mixed distribution of the cross-domain data and generate the prediction sequence for the target domain directly utilizing conditional sampling. The proposed TimeControl contains three pivotal designs: (1) The condition network captures the multi-scale fluctuation patterns from the observation sequence, which are utilized as context representations to guide the denoising network to generate the prediction sequence; (2) Adapter-based fine-tuning strategy, the multi-domain universal representation learned in the pretraining stage is utilized for downstream tasks in target domains; (3) A novel hybrid architecture is designed to align the observation and prediction spaces, enabling TimeControl to generate prediction sequences of arbitrary lengths with flexibility. We conduct extensive experiments on mainstream 49 benchmarks and 30 baselines, and the TimeControl outperforms existing baselines on all data domains, exhibiting superior zero-shot generalization ability.

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。

↑