DC-Mamber: A Dual Channel Prediction Model based on Mamba and Linear Transformer for Multivariate Time Series Forecasting

2025-07-06

AI Summary

To address the challenge of jointly modeling local temporal patterns and global inter-variable dependencies in multivariate time series forecasting (MTSF), this paper proposes DC-Mamba, a dual-channel forecasting framework. It innovatively integrates channel-independent Mamba—capable of capturing long-range univariate temporal dynamics—with channel-mixed linear Transformer—efficiently modeling cross-variable global dependencies—via a dual-encoder architecture and a collaborative embedding-fusion mechanism. Unlike standard Transformers (high computational complexity, weak local sensitivity) and vanilla Mamba (inherently limited in parallel global context modeling), DC-Mamba retains linear time complexity while substantially enhancing both local pattern sensitivity and global contextual aggregation. Extensive experiments on eight public benchmarks demonstrate that DC-Mamba achieves state-of-the-art prediction accuracy, outperforming leading methods across diverse domains. Moreover, it exhibits superior computational efficiency and strong generalization capability.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

In multivariate time series forecasting (MTSF), existing strategies for processing sequences are typically categorized as channel-independent and channel-mixing. The former treats all temporal information of each variable as a token, focusing on capturing local temporal features of individual variables, while the latter constructs a token from the multivariate information at each time step, emphasizing the modeling of global temporal dependencies. Current mainstream models are mostly based on Transformer and the emerging Mamba. Transformers excel at modeling global dependencies through self-attention mechanisms but exhibit limited sensitivity to local temporal patterns and suffer from quadratic computational complexity, restricting their efficiency in long-sequence processing. In contrast, Mamba, based on state space models (SSMs), achieves linear complexity and efficient long-range modeling but struggles to aggregate global contextual information in parallel. To overcome the limitations of both models, we propose DC-Mamber, a dual-channel forecasting model based on Mamba and linear Transformer for time series forecasting. Specifically, the Mamba-based channel employs a channel-independent strategy to extract intra-variable features, while the Transformer-based channel adopts a channel-mixing strategy to model cross-timestep global dependencies. DC-Mamber first maps the raw input into two distinct feature representations via separate embedding layers. These representations are then processed by a variable encoder (built on Mamba) and a temporal encoder (built on linear Transformer), respectively. Finally, a fusion layer integrates the dual-channel features for prediction. Extensive experiments on eight public datasets confirm DC-Mamber's superior accuracy over existing models.

Problem

• Overcoming limitations in multivariate time series forecasting models
• Balancing local and global temporal feature extraction
• Improving efficiency and accuracy in long-sequence processing

Innovation

• Dual-channel model combines Mamba and Transformer
• Mamba extracts intra-variable temporal features
• Linear Transformer models cross-timestep dependencies