N-BEATS-RNN: deep learning for time series forecasting

This work presents N-BEATS-RNN, an extended version of an existing ensemble of deep learning networks for time series forecasting, N-BEATS. We apply a state-of-the-art Neural Architecture Search, based on a fast and efficient weight-sharing search, to solve for an ideal Recurrent Neural Network arch...

Full description

Saved in:
Bibliographic Details
Published in2020 19th IEEE International Conference on Machine Learning and Applications (ICMLA) pp. 765 - 768
Main Authors Sbrana, Attilio, Debiaso Rossi, Andre Luis, Coelho Naldi, Murilo
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2020
Subjects
Deep learning
Forecasting
M4 competition
neural architecture search
Predictive models
Recurrent neural networks
Time series analysis
Time series forecasting
Time-frequency analysis
Training
weight sharing
Online AccessGet full text
DOI10.1109/ICMLA51294.2020.00125

Cover

More Information
Summary:This work presents N-BEATS-RNN, an extended version of an existing ensemble of deep learning networks for time series forecasting, N-BEATS. We apply a state-of-the-art Neural Architecture Search, based on a fast and efficient weight-sharing search, to solve for an ideal Recurrent Neural Network architecture to be added to N-BEATS. We evaluated the proposed N-BEATS-RNN architecture in the widely-known M4 competition dataset, which contains 100,000 time series from a variety of sources. N-BEATS-RNN achieves comparable results to N-BEATS and the M4 competition winner while employing solely 108 models, as compared to the original 2,160 models employed by N-BEATS, when composing its final ensemble of forecasts. Thus, N-BEATS-RNN's biggest contribution is in its training time reduction, which is in the order of 9x compared with the original ensembles in N-BEATS.
DOI:10.1109/ICMLA51294.2020.00125