CoRA: 通过相关性感知适配器提升多变量预测的时序基础模型
CoRA: Boosting Time Series Foundation Models for Multivariate Forecasting through Correlation-aware Adapter
AI总结：
本文提出CoRA，一种轻量级插件式方法，通过捕捉不同类型的相关性提升多变量时间序列预测性能，采用低秩时间变化和时间不变成分分解相关矩阵，并引入双对比学习方法。
摘要：
大多数现有时间序列基础模型（TSFMs）采用通道独立建模，专注于捕捉和泛化时间依赖性，而忽视通道间的相关性或忽略相关性的不同方面。然而，这些相关性在多变量时间序列预测中起着重要作用。为此，我们提出了一种相关性感知适配器（CoRA），一种轻量级插件式方法，仅需通过TSFMs进行微调即可捕捉不同类型的相关性，从而提升预测性能。具体而言，为减少复杂度，我们创新性地将相关矩阵分解为低秩时间变化和时间不变成分。对于时间变化成分，我们进一步设计可学习的多项式来学习动态相关性，通过捕捉趋势或周期性模式。为了学习仅出现在某些通道之间的正负相关性，我们引入了一种新颖的双对比学习方法，通过投影层识别相关性，由异构部分对比损失在训练过程中调节，无需在推理阶段引入额外复杂度。在10个真实世界数据集上的广泛实验表明，CoRA可以提升TSFMs在多变量预测性能上的表现。
英文摘要：
Most existing Time Series Foundation Models (TSFMs) use channel independent modeling and focus on capturing and generalizing temporal dependencies, while neglecting the correlations among channels or overlooking the different aspects of correlations. However, these correlations play a vital role in Multivariate time series forecasting. To address this, we propose a CoRrelation-aware Adapter (CoRA), a lightweight plug-and-play method that requires only fine-tuning with TSFMs and is able to capture different types of correlations, so as to improve forecast performance. Specifically, to reduce complexity, we innovatively decompose the correlation matrix into low-rank Time-Varying and Time-Invariant components. For the Time-Varying component, we further design learnable polynomials to learn dynamic correlations by capturing trends or periodic patterns. To learn positive and negative correlations that appear only among some channels, we introduce a novel dual contrastive learning method that identifies correlations through projection layers, regulated by a Heterogeneous-Partial contrastive loss during training, without introducing additional complexity in the inference stage. Extensive experiments on 10 real-world datasets demonstrate that CoRA can improve TSFMs in multivariate forecasting performance.
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。
