Temporal Pattern Attention for Multivariate Time Series Forecasting

TL;DR

This paper introduces a novel attention-based model that uses frequency domain filtering to capture long-term temporal patterns in multivariate time series, significantly improving forecasting accuracy.

Contribution

The paper proposes a new temporal pattern attention mechanism that leverages frequency domain filtering to better model long-term dependencies in multivariate time series forecasting.

Findings

Achieved state-of-the-art results on multiple real-world datasets.

Effectively captures long-term dependencies using frequency domain features.

Outperforms traditional attention mechanisms in forecasting accuracy.

Abstract

Forecasting multivariate time series data, such as prediction of electricity consumption, solar power production, and polyphonic piano pieces, has numerous valuable applications. However, complex and non-linear interdependencies between time steps and series complicate the task. To obtain accurate prediction, it is crucial to model long-term dependency in time series data, which can be achieved to some good extent by recurrent neural network (RNN) with attention mechanism. Typical attention mechanism reviews the information at each previous time step and selects the relevant information to help generate the outputs, but it fails to capture the temporal patterns across multiple time steps. In this paper, we propose to use a set of filters to extract time-invariant temporal patterns, which is similar to transforming time series data into its "frequency domain". Then we proposed a novel attention mechanism to select relevant time series, and use its "frequency domain" information for forecasting. We applied the proposed model on several real-world tasks and achieved state-of-the-art performance in all of them with only one exception.