Tailored Architectures for Time Series Forecasting: Evaluating Deep Learning Models on Gaussian Process-Generated Data

TL;DR

This paper introduces a Gaussian Process-generated dataset with known characteristics and a new modular model, TimeFlex, to evaluate deep learning architectures for time series forecasting across diverse data properties.

Contribution

The study presents a novel dataset with explicit characteristics and a new adaptable model, TimeFlex, to better understand model performance relative to data features.

Findings

TimeFlex outperforms existing models on diverse time series characteristics.

The Gaussian Process dataset reveals strengths and weaknesses of different architectures.

Model performance varies significantly with data properties.

Abstract

Developments in Deep Learning have significantly improved time series forecasting by enabling more accurate modeling of complex temporal dependencies inherent in sequential data. The effectiveness of such models is often demonstrated on limited sets of specific real-world data. Although this allows for comparative analysis, it still does not demonstrate how specific data characteristics align with the architectural strengths of individual models. Our research aims at uncovering clear connections between time series characteristics and particular models. We introduce a novel dataset generated using Gaussian Processes, specifically designed to display distinct, known characteristics for targeted evaluations of model adaptability to them. Furthermore, we present TimeFlex, a new model that incorporates a modular architecture tailored to handle diverse temporal dynamics, including trends and periodic patterns. This model is compared to current state-of-the-art models, offering a deeper understanding of how models perform under varied time series conditions.