Deep Autoregressive Models with Spectral Attention

TL;DR

This paper introduces a novel deep autoregressive forecasting model that incorporates spectral attention to effectively capture global and local frequency domain information, improving accuracy and interpretability.

Contribution

The paper presents Spectral Attention Autoregressive Model (SAAM), combining spectral filtering with autoregressive models for enhanced time series forecasting.

Findings

SAAM outperforms state-of-the-art models on multiple datasets.

The model requires fewer parameters and offers interpretability.

Spectral filtering effectively captures trends and seasonality.

Abstract

Time series forecasting is an important problem across many domains, playing a crucial role in multiple real-world applications. In this paper, we propose a forecasting architecture that combines deep autoregressive models with a Spectral Attention (SA) module, which merges global and local frequency domain information in the model's embedded space. By characterizing in the spectral domain the embedding of the time series as occurrences of a random process, our method can identify global trends and seasonality patterns. Two spectral attention models, global and local to the time series, integrate this information within the forecast and perform spectral filtering to remove time series's noise. The proposed architecture has a number of useful properties: it can be effectively incorporated into well-know forecast architectures, requiring a low number of parameters and producing interpretable results that improve forecasting accuracy. We test the Spectral Attention Autoregressive Model (SAAM) on several well-know forecast datasets, consistently demonstrating that our model compares favorably to state-of-the-art approaches.