GAMMA-Net: Adaptive Long-Horizon Traffic Spatio-Temporal Forecasting Model based on Interleaved Graph Attention and Multi-Axis Mamba

2026-04-18

AI Summary

Existing traffic forecasting methods struggle to effectively model complex long-range spatiotemporal dependencies. This work proposes a novel architecture that integrates Graph Attention Networks (GAT) with a multi-axis Mamba state space model, uniquely intertwining multi-axis selective state space mechanisms with graph attention to adaptively capture dynamic spatial correlations and efficiently model long-term temporal dependencies. The proposed method achieves significant performance gains over state-of-the-art models across multiple benchmark datasets, reducing the Mean Absolute Error (MAE) by up to 16.25%. These results demonstrate its superior predictive accuracy and computational efficiency across varying forecasting horizons.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Accurate traffic forecasting is crucial for intelligent transportation systems, supporting effective traffic management, congestion reduction, and informed urban planning. However, traditional models often fail to adequately capture the intricate spatio-temporal dependencies present in traffic data. To overcome these limitations, we introduce GAMMA-Net, a novel approach that integrates Graph Attention Networks (GAT) with multi-axis Selective State Space Models (Mamba). The GAT component uses a self-attention mechanism to dynamically adjust the influence of nodes within the traffic network, enabling adaptive spatial dependency modeling based on real-time conditions. Simultaneously, the Mamba module efficiently models long-term temporal and spatial dynamics without the heavy computational cost of conventional recurrent architectures. Extensive experiments on several benchmark traffic datasets, including METR-LA, PEMS-BAY, PEMS03, PEMS04, PEMS07, and PEMS08, show that GAMMA-Net consistently outperforms existing state-of-the-art models across different prediction horizons, achieving up to a 16.25% reduction in Mean Absolute Error (MAE) compared to baseline models. Ablation studies highlight the critical contributions of both the spatial and temporal components, emphasizing their complementary role in improving prediction accuracy. In conclusion, the GAMMA-Net model sets a new standard in traffic forecasting, offering a powerful tool for next-generation traffic management and urban planning. The code for this study is available at https://github.com/hdy6438/GAMMA-Net

Problem

• traffic forecasting
• spatio-temporal dependencies
• long-horizon prediction
• intelligent transportation systems

Innovation

• Graph Attention Network
• Mamba
• Spatio-Temporal Forecasting
• Long-Horizon Prediction
• Selective State Space Model