DUET: Dual Clustering Enhanced Multivariate Time Series Forecasting

TL;DR

DUET addresses two core challenges in multivariate time series forecasting: heterogeneous temporal patterns from Temporal Distribution Shift and complex, intertwined channel relationships. It introduces a Temporal Clustering Module to partition time series into distributions and apply distribution-specific pattern extractors, and a Channel Clustering Module that learns frequency-domain channel relations with sparsification via a mask. The Fusion Module combines these components through a masked attention mechanism, enabling robust forecasting. Extensive experiments on 25 real-world datasets across 10 domains demonstrate state-of-the-art performance, highlighting DUET's effectiveness in handling temporal heterogeneity and channel noise, with publicly available code for reproducibility.

Abstract

Multivariate time series forecasting is crucial for various applications, such as financial investment, energy management, weather forecasting, and traffic optimization. However, accurate forecasting is challenging due to two main factors. First, real-world time series often show heterogeneous temporal patterns caused by distribution shifts over time. Second, correlations among channels are complex and intertwined, making it hard to model the interactions among channels precisely and flexibly. In this study, we address these challenges by proposing a general framework called DUET, which introduces dual clustering on the temporal and channel dimensions to enhance multivariate time series forecasting. First, we design a Temporal Clustering Module (TCM) that clusters time series into fine-grained distributions to handle heterogeneous temporal patterns. For different distribution clusters, we design various pattern extractors to capture their intrinsic temporal patterns, thus modeling the heterogeneity. Second, we introduce a novel Channel-Soft-Clustering strategy and design a Channel Clustering Module (CCM), which captures the relationships among channels in the frequency domain through metric learning and applies sparsification to mitigate the adverse effects of noisy channels. Finally, DUET combines TCM and CCM to incorporate both the temporal and channel dimensions. Extensive experiments on 25 real-world datasets from 10 application domains, demonstrate the state-of-the-art performance of DUET.

Figures (11)

• Figure 1: A Non-stationary time series with three intervals A, B, and C, exhibiting varying value distributions ($P_A \ne P_B \ne P_C$) and temporal patterns.
• Figure 2: Channel strategies. Different colors represent different channels, with squares representing features before processing with various channel strategies, and squares with rounded corners representing features after processing.
• Figure 3: Performance of DUET. Results (MSE) are averaged from all forecasting horizons. DUET outperforms strong baselines in 10 commonly used datasets.
• Figure 4: The architecture of DUET. Temporal Clustering Module clusters time series into fine-grained distribution. For different distribution clusters, various pattern extractors are designed to capture their intrinsic temporal patterns. Channel Clustering Module flexibly captures the relationships among channels in the frequency domain space through Metric Learning and applies Sparsification. Fusion Module combines the temporal features and the channel mask matrix.
• Figure 5: (a) The structure of the Distribution Router, which consists of Distribution Characterization and Routing. (b) The structure of the Linear-based Pattern Extractor, which decomposes the series into seasonal and trend parts, separately extracts temporal features with linear models and reads them. (c) The Learnable Distance Metric is to capture the relationships among channels. (d) The Fusion Module is to combine the temporal features and the channel mask matrix.

Theorems & Definitions (2)

• definition 1: Time series
• definition 2: Temporal Distribution Shift du2021adarnn