Decomposing the Time Series Forecasting Pipeline: A Modular Approach for Time Series Representation, Information Extraction, and Projection

TL;DR

This paper proposes a modular pipeline for time series forecasting that decomposes the task into three stages, enabling systematic evaluation of architectural choices and achieving state-of-the-art accuracy with improved efficiency.

Contribution

It introduces a modular framework for time series forecasting, systematically analyzing architectural components across diverse datasets, and demonstrating improved accuracy and efficiency.

Findings

Achieves state-of-the-art forecasting accuracy.

Reduces training and inference times.

Uses fewer parameters than previous models.

Abstract

With the advent of Transformers, time series forecasting has seen significant advances, yet it remains challenging due to the need for effective sequence representation, memory construction, and accurate target projection. Time series forecasting remains a challenging task, demanding effective sequence representation, meaningful information extraction, and precise future projection. Each dataset and forecasting configuration constitutes a distinct task, each posing unique challenges the model must overcome to produce accurate predictions. To systematically address these task-specific difficulties, this work decomposes the time series forecasting pipeline into three core stages: input sequence representation, information extraction and memory construction, and final target projection. Within each stage, we investigate a range of architectural configurations to assess the effectiveness of various modules, such as convolutional layers for feature extraction and self-attention mechanisms for information extraction, across diverse forecasting tasks, including evaluations on seven benchmark datasets. Our models achieve state-of-the-art forecasting accuracy while greatly enhancing computational efficiency, with reduced training and inference times and a lower parameter count. The source code is available at https://github.com/RobertLeppich/REP-Net.