Journal
PATTERN RECOGNITION
Volume 120, Issue -, Pages -Publisher
ELSEVIER SCI LTD
DOI: 10.1016/j.patcog.2021.108148
Keywords
Time series forecasting; Global forecasting models; Data augmentation; Transfer learning; RNN
Funding
- Australian Research Council [DE190100045]
- Facebook Statistics for Improving Insights and Decisions research award
- Monash Institute of Medical Engineering
- MASSIVE-High performance computing facility, Australia
- National Natural Science Foundation of China [11701022]
- Australian Research Council [DE190100045] Funding Source: Australian Research Council
Ask authors/readers for more resources
Global Forecasting Models (GFM) trained across multiple time series have shown promising results in forecasting competitions and real-world applications. A novel data-augmentation based framework has been proposed to improve the accuracy of GFM models in less data-abundant settings, using techniques such as GRATIS, MBB, and DBA. The proposed variants significantly outperform state-of-the-art univariate forecasting methods in competition and real-world datasets.
Forecasting models that are trained across sets of many time series, known as Global Forecasting Models (GFM), have shown recently promising results in forecasting competitions and real-world applications, outperforming many state-of-the-art univariate forecasting techniques. In most cases, GFMs are implemented using deep neural networks, and in particular Recurrent Neural Networks (RNN), which require a sufficient amount of time series to estimate their numerous model parameters. However, many time series databases have only a limited number of time series. In this study, we propose a novel, data augmentation based forecasting framework that is capable of improving the baseline accuracy of the GFM models in less data-abundant settings. We use three time series augmentation techniques: GRATIS, moving block bootstrap (MBB), and dynamic time warping barycentric averaging (DBA) to synthetically generate a collection of time series. The knowledge acquired from these augmented time series is then transferred to the original dataset using two different approaches: the pooled approach and the transfer learning approach. When building GFMs, in the pooled approach, we train a model on the augmented time series alongside the original time series dataset, whereas in the transfer learning approach, we adapt a pre trained model to the new dataset. In our evaluation on competition and real-world time series datasets, our proposed variants can significantly improve the baseline accuracy of GFM models and outperform state-of-the-art univariate forecasting methods. (c) 2021 Elsevier Ltd. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available