An evolutionary neural architecture search-based approach for time series forecasting.

Vu, Trung Hieu; Nguyen, Tien Thanh; Elyan, Eyad

Abstract

Time series forecasting (TSF) is one of the most prevalent research topics in artificial intelligence and has garnered significant attention in the research community. In recent years, significant breakthroughs have been made in TSF research, shifting from traditional statistical models to deep learning (DL), and attention-based methods. While attention-based methods excel at capturing global dependencies, they often face challenges in effectively modeling local patterns. Additionally, the design of these networks typically demands substantial human expertise, experimental work, and manual configuration. To address these issues, we propose an Evolutionary Neural Architecture Search for Time Series Forecasting, entitled ENAS-TSF to automate TSF architecture design. Concretely, we propose a novel local/ global context module encoding strategy to define a search space. Each local context module includes various convolutions with different kernel sizes to capture temporal dependencies, aiming to enhance the local features and pattern recognition. Global encoding meanwhile contains attention mechanisms and feedforward layers for global context modeling. We propose an evolutionary neural architecture search approach to identify the optimal ENAS-TSF architectures, achieving the ideal balance of local/ global context modeling. Extensive experiments on common benchmark datasets show that ENAS-TSF achieves competitive performance compared to state-of-the-art methods, demonstrating the proposed framework's effectiveness.

Citation

VU, T.H., NGUYEN, T.T. and ELYAN, E. 2025. An evolutionary neural architecture search-based approach for time series forecasting. In Proceedings of the 2025 IEEE (Institute of Electrical and Electronics Engineers) Congress on evolutionary computation (CEC 2025), 8-12 June 2025, Hangzhou, China. Piscataway: IEEE [online], article number 11043002, pages 1-8. Available from: https://doi.org/10.1109/cec65147.2025.11043002

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2025 IEEE (Institute of Electrical and Electronics Engineers) Congress on evolutionary computation (CEC 2025)
• Start Date: Jun 8, 2025
• End Date: Jun 12, 2025
• Acceptance Date: Jan 31, 2025
• Online Publication Date: Jun 8, 2025
• Publication Date: Jun 30, 2025
• Deposit Date: Jun 27, 2025
• Publicly Available Date: Jun 27, 2025
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Article Number: 11043002
• Pages: 1-8
• DOI: https://doi.org/10.1109/cec65147.2025.11043002
• Keywords: Time series forecasting; Neural architecture search; Deep learning; Evolutionary algorithm
• Public URL: https://rgu-repository.worktribe.com/output/2892654