MLGN: multi-scale local-global feature learning network for long-term series forecasting
Open Access
- 1 December 2023
- journal article
- research article
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 4 (4) , 045059
Abstract
Although Transformer-based methods have achieved remarkable performance in the field of long-term series forecasting, they can be computationally expensive and lack the ability to specifically model local features as CNNs. CNN-based methods, such as temporal convolutional network (TCN), utilize convolutional filters to capture local temporal features. However, the intermediate layers of TCN suffer from a limited effective receptive field, which can result in the loss of temporal relations during global feature extraction.To solve the above problems, we propose to combine local features and global correlations to capture the overall view of time series (e.g. fluctuations, trends). To fully exploit the underlying information in the time series, a multi-scale branch structure is adopted to model different potential patterns separately. Each pattern is extracted using a combination of interactive learning convolution and causal frequency enhancement to capture both local features and global correlations. Furthermore, our proposed method,multi-scale local-global feature learning network (MLGN), achieves a time and memory complexity of O(L) and consistently achieve state-of-the-art results on six benchmark datasets. In comparision with previous best method Fedformer, MLGN yields 12.98% and 11.38% relative improvements for multivariate and univariate time series, respectively. Our code and data are available on Github at https://github.com/Zero-coder/MLGN.Keywords
Funding Information
- National Natural Science Foundation of China (62073313)
- The National Key R&D Program of China (2020YFB1709702)
This publication has 26 references indexed in Scilit:
- Implementing transfer learning across different datasets for time series forecastingPattern Recognition, 2021
- A new ensemble deep learning approach for exchange rates forecasting and tradingAdvanced Engineering Informatics, 2020
- A CNN–LSTM model for gold price time-series forecastingNeural Computing and Applications, 2020
- Multivariate Temporal Convolutional Network: A Deep Neural Networks Approach for Multivariate Time Series ForecastingElectronics, 2019
- A vision-based statistical methodology for automatically modeling continuous urban traffic flowsAdvanced Engineering Informatics, 2018
- Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networks2018
- A novel approach for precipitation forecast via improved K-nearest neighbor algorithmAdvanced Engineering Informatics, 2017
- Assessment and weighting of meteorological ensemble forecast members based on supervised machine learning with application to runoff simulations and flood warningAdvanced Engineering Informatics, 2017
- DIRECT MULTI‐STEP ESTIMATION AND FORECASTINGJournal of Economic Surveys, 2007
- A hybrid SARIMA and support vector machines in forecasting the production values of the machinery industry in TaiwanExpert Systems with Applications, 2007