EconBase
← All papers

W-Transformers : A Wavelet-based Transformer Framework for Univariate Time Series Forecasting

Lena Sasal, Tanujit Chakraborty, Abdenour Hadid

arXiv 8 Sep 2022 · Machine Learning · 40 citations (OpenAlex)

arXiv:2209.03945 · PDF · DOI · OpenAlex · Extracted main text

Abstract

Deep learning utilizing transformers has recently achieved a lot of success in many vital areas such as natural language processing, computer vision, anomaly detection, and recommendation systems, among many others. Among several merits of transformers, the ability to capture long-range temporal dependencies and interactions is desirable for time series forecasting, leading to its progress in various time series applications. In this paper, we build a transformer model for non-stationary time series. The problem is challenging yet crucially important. We present a novel framework for univariate time series representation learning based on the wavelet-based transformer encoder architecture and call it W-Transformer. The proposed W-Transformers utilize a maximal overlap discrete wavelet transformation (MODWT) to the time series data and build local transformers on the decomposed datasets to vividly capture the nonstationarity and long-range nonlinear dependencies in the time series. Evaluating our framework on several publicly available benchmark time series datasets from various domains and with diverse characteristics, we demonstrate that it performs, on average, significantly better than the baseline forecasters for short-term and long-term forecasting, even for datasets that consist of only a few hundred training samples.

Citation extraction

48
references
88
in-text mentions
48
distinct cited
0
self-citations
4,776
main-text words

appendix boundary found by none_found · 100% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1D. B. Percival and A. T. Walden, Wavelet methods for time series ana… (2000) vol. 41.00063100%
2M. Panja, T. Chakraborty, U. Kumar, and N. Liu, “Epicasting: An ense… (2022) Epicasting: An ensemble wavelet neural network (ewnet) for forecasting epidemics1.00053100%
3A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gom… (2017) Attention is all you need1.00053100%
4R. Hyndman, A. B. Koehler, J. K. Ord, and R. D. Snyder, Forecasting… (2008)0.81142100%
5M. Aminghafari and J.-M. Poggi, “Forecasting time series using wavel… (2007) Forecasting time series using wavelets0.73732100%
6J. Faraway and C. Chatfield, “Time series forecasting with neural ne… (1998) Time series forecasting with neural networks: a comparative study using the air line data0.73732100%
7R. J. Hyndman and G. Athanasopoulos, Forecasting: principles and pra… (2018)0.73732100%
8D. Salinas, V. Flunkert, J. Gasthaus, and T. Januschowski, “Deepar:… (2020) Deepar: Probabilistic forecasting with autoregressive recurrent networks0.73732100%
9G. Zerveas, S. Jayaraman, D. Patel, A. Bhamidipaty, and C. Eickhoff,… (2021) A transformer-based framework for multivariate time series representation learning0.73732100%
10R. Godahewa, C. Bergmeir, G. I. Webb, R. J. Hyndman, and P. Montero-… (2021) Monash time series forecasting archive0.64422100%

Showing the top 10 of 48 scored citations.