Frequency-domain MLPs are More Effective Learners in Time Series Forecasting
Preprint
Abstract
Time series forecasting has played the key role in different industrial, including finance, traffic, energy, and healthcare domains. While existing literatures have designed many sophisticated architectures based on RNNs, GNNs, or Transformers, another kind of approaches based on multi-layer perceptrons (MLPs) are proposed with simple structure, low complexity, and {superior performance}. However, most MLP-based forecasting methods suffer from the point-wise mappings and information bottleneck, which largely hinders the forecasting performance. To overcome this problem, we explore a novel direction of applying MLPs in the frequency domain for time series forecasting. We investigate the learned patterns of frequency-domain MLPs and discover their two inherent characteristic benefiting forecasting, (i) global view: frequency spectrum makes MLPs own a complete view for signals and learn global dependencies more easily, and (ii) energy compaction: frequency-domain MLPs concentrate on smaller key part of frequency components with compact signal energy. Then, we propose FreTS, a simple yet effective architecture built upon Frequency-domain MLPs for Time Series forecasting. FreTS mainly involves two stages, (i) Domain Conversion, that transforms time-domain signals into complex numbers of frequency domain; (ii) Frequency Learning, that performs our redesigned MLPs for the learning of real and imaginary part of frequency components. The above stages operated on both inter-series and intra-series scales further contribute to channel-wise and time-wise dependency learning. Extensive experiments on 13 real-world benchmarks (including 7 benchmarks for short-term forecasting and 6 benchmarks for long-term forecasting) demonstrate our consistent superiority over state-of-the-art methods.Keywords
All Related Versions
This publication has 17 references indexed in Scilit:
- A Survey on Deep Learning based Time Series Analysis with Frequency Transformation2025
- Learning Informative Representation for Fairness-Aware Multivariate Time-Series Forecasting: A Group-Based PerspectiveIEEE Transactions on Knowledge and Data Engineering, 2023
- Informer: Beyond Efficient Transformer for Long Sequence Time-Series ForecastingProceedings of the AAAI Conference on Artificial Intelligence, 2021
- Time-series forecasting with deep learning: a surveyPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2021
- Deep Learning for Time Series Forecasting: A SurveyBig Data, 2021
- Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks2020
- DeepAR: Probabilistic forecasting with autoregressive recurrent networksInternational Journal of Forecasting, 2020
- Stock Price Prediction via Discovering Multi-Frequency Trading Patterns2017
- ARIMA Models and the Box-Jenkins Methodology2016
- Forecasting Fine-Grained Air Quality Based on Big Data2015