Should We Reconsider RNNs for Time-Series Forecasting?

(1) Background: In recent years, Transformer-based models have dominated the time-series forecasting domain, overshadowing recurrent neural networks (RNNs) such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). While Transformers demonstrate superior performance, their high computatio...

Full description

Saved in:
Bibliographic Details
Main Authors: Vahid Naghashi, Mounir Boukadoum, Abdoulaye Banire Diallo
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:AI
Subjects:
Online Access:https://www.mdpi.com/2673-2688/6/5/90
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849711416074829824
author Vahid Naghashi
Mounir Boukadoum
Abdoulaye Banire Diallo
author_facet Vahid Naghashi
Mounir Boukadoum
Abdoulaye Banire Diallo
author_sort Vahid Naghashi
collection DOAJ
description (1) Background: In recent years, Transformer-based models have dominated the time-series forecasting domain, overshadowing recurrent neural networks (RNNs) such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). While Transformers demonstrate superior performance, their high computational cost limits their practical application in resource-constrained settings. (2) Methods: In this paper, we reconsider RNNs—specifically the GRU architecture—as an efficient alternative to time-series forecasting by leveraging this architecture’s sequential representation capability to capture cross-channel dependencies effectively. Our model also utilizes a feed-forward layer right after the GRU module to represent temporal dependencies, and aggregates it with the GRU layers to predict future values of a given time-series. (3) Results and conclusions: Our extensive experiments conducted on different real-world datasets show that our inverted GRU (iGRU) model achieves promising results in terms of error metrics and memory efficiency, challenging or surpassing state-of-the-art models on various benchmarks.
format Article
id doaj-art-592b0b08a46348b099ba3c21167f4b6f
institution DOAJ
issn 2673-2688
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series AI
spelling doaj-art-592b0b08a46348b099ba3c21167f4b6f2025-08-20T03:14:38ZengMDPI AGAI2673-26882025-04-01659010.3390/ai6050090Should We Reconsider RNNs for Time-Series Forecasting?Vahid Naghashi0Mounir Boukadoum1Abdoulaye Banire Diallo2Department of Computer Science, Université du Québec à Montréal, Montreal, QC H2L 2C4, CanadaDepartment of Computer Science, Université du Québec à Montréal, Montreal, QC H2L 2C4, CanadaDepartment of Computer Science, Université du Québec à Montréal, Montreal, QC H2L 2C4, Canada(1) Background: In recent years, Transformer-based models have dominated the time-series forecasting domain, overshadowing recurrent neural networks (RNNs) such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). While Transformers demonstrate superior performance, their high computational cost limits their practical application in resource-constrained settings. (2) Methods: In this paper, we reconsider RNNs—specifically the GRU architecture—as an efficient alternative to time-series forecasting by leveraging this architecture’s sequential representation capability to capture cross-channel dependencies effectively. Our model also utilizes a feed-forward layer right after the GRU module to represent temporal dependencies, and aggregates it with the GRU layers to predict future values of a given time-series. (3) Results and conclusions: Our extensive experiments conducted on different real-world datasets show that our inverted GRU (iGRU) model achieves promising results in terms of error metrics and memory efficiency, challenging or surpassing state-of-the-art models on various benchmarks.https://www.mdpi.com/2673-2688/6/5/90time-seriesgated recurrent unitstemporal dependenciescross-channel correlations
spellingShingle Vahid Naghashi
Mounir Boukadoum
Abdoulaye Banire Diallo
Should We Reconsider RNNs for Time-Series Forecasting?
AI
time-series
gated recurrent units
temporal dependencies
cross-channel correlations
title Should We Reconsider RNNs for Time-Series Forecasting?
title_full Should We Reconsider RNNs for Time-Series Forecasting?
title_fullStr Should We Reconsider RNNs for Time-Series Forecasting?
title_full_unstemmed Should We Reconsider RNNs for Time-Series Forecasting?
title_short Should We Reconsider RNNs for Time-Series Forecasting?
title_sort should we reconsider rnns for time series forecasting
topic time-series
gated recurrent units
temporal dependencies
cross-channel correlations
url https://www.mdpi.com/2673-2688/6/5/90
work_keys_str_mv AT vahidnaghashi shouldwereconsiderrnnsfortimeseriesforecasting
AT mounirboukadoum shouldwereconsiderrnnsfortimeseriesforecasting
AT abdoulayebanirediallo shouldwereconsiderrnnsfortimeseriesforecasting