Temporal Query Network for Efficient Multivariate Time Series Forecasting

TL;DR

This paper introduces Temporal Query Network (TQNet), a novel attention-based model that effectively captures multivariate correlations in time series data, achieving state-of-the-art forecasting accuracy with high efficiency.

Contribution

The paper proposes the Temporal Query (TQ) technique and TQNet, a lightweight model that improves multivariate time series forecasting by better modeling variable correlations.

Findings

TQNet achieves state-of-the-art accuracy on 12 datasets.

TQNet balances performance and computational efficiency.

TQNet outperforms existing methods in multivariate correlation modeling.

Abstract

Sufficiently modeling the correlations among variables (aka channels) is crucial for achieving accurate multivariate time series forecasting (MTSF). In this paper, we propose a novel technique called Temporal Query (TQ) to more effectively capture multivariate correlations, thereby improving model performance in MTSF tasks. Technically, the TQ technique employs periodically shifted learnable vectors as queries in the attention mechanism to capture global inter-variable patterns, while the keys and values are derived from the raw input data to encode local, sample-level correlations. Building upon the TQ technique, we develop a simple yet efficient model named Temporal Query Network (TQNet), which employs only a single-layer attention mechanism and a lightweight multi-layer perceptron (MLP). Extensive experiments demonstrate that TQNet learns more robust multivariate correlations, achieving state-of-the-art forecasting accuracy across 12 challenging real-world datasets. Furthermore, TQNet achieves high efficiency comparable to linear-based methods even on high-dimensional datasets, balancing performance and computational cost. The code is available at: https://github.com/ACAT-SCUT/TQNet.