HGTS-Former: Hierarchical HyperGraph Transformer for Multivariate Time Series Analysis

2025-08-04

AI Summary

To address the challenges of modeling dynamic evolution and high-order inter-variable couplings in high-dimensional multivariate time series, this paper proposes HGTS-Former—the first hierarchical Transformer-based model integrating hypergraph structural priors. Its core innovation lies in constructing a dynamic hierarchical hypergraph that explicitly captures intra-channel temporal patterns and fine-grained, high-order cross-variable dependencies; an EdgeToNode module is further introduced to enable learnable mapping from hyperedge features to node representations. The architecture integrates time-block embeddings, multi-head self-attention, and feed-forward enhancement. Extensive experiments across two task categories and eight benchmark datasets demonstrate that HGTS-Former significantly outperforms state-of-the-art methods, validating its superior capability in modeling complex temporal dependencies and strong generalizability.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Multivariate time series analysis has long been one of the key research topics in the field of artificial intelligence. However, analyzing complex time series data remains a challenging and unresolved problem due to its high dimensionality, dynamic nature, and complex interactions among variables. Inspired by the strong structural modeling capability of hypergraphs, this paper proposes a novel hypergraph-based time series transformer backbone network, termed HGTS-Former, to address the multivariate coupling in time series data. Specifically, given the multivariate time series signal, we first normalize and embed each patch into tokens. Then, we adopt the multi-head self-attention to enhance the temporal representation of each patch. The hierarchical hypergraphs are constructed to aggregate the temporal patterns within each channel and fine-grained relations between different variables. After that, we convert the hyperedge into node features through the EdgeToNode module and adopt the feed-forward network to further enhance the output features. Extensive experiments conducted on two multivariate time series tasks and eight datasets fully validated the effectiveness of our proposed HGTS-Former. The source code will be released on https://github.com/Event-AHU/Time_Series_Analysis.

Problem

• Analyzing high-dimensional dynamic multivariate time series
• Modeling complex interactions among time series variables
• Enhancing temporal representation with hierarchical hypergraphs

Innovation

• Hierarchical hypergraphs model complex variable relations
• Multi-head self-attention enhances temporal representation
• EdgeToNode converts hyperedge features into nodes