Numerion: A Multi-Hypercomplex Model for Time Series Forecasting
Abstract
Many methods aim to enhance time series forecasting by decomposing the series through intricate model structures and prior knowledge, yet they are inevitably limited by computational complexity and the robustness of the assumptions. Our research uncovers that in the complex domain and higher-order hypercomplex spaces, the characteristic frequencies of time series naturally decrease. Leveraging this insight, we propose Numerion, a time series forecasting model based on multiple hypercomplex spaces. Specifically, grounded in theoretical support, we generalize linear layers and activation functions to hypercomplex spaces of arbitrary power-of-two dimensions and introduce a novel Real-Hypercomplex-Real Domain Multi-Layer Perceptron (RHR-MLP) architecture. Numerion utilizes multiple RHR-MLPs to map time series into hypercomplex spaces of varying dimensions, naturally decomposing and independently modeling the series, and adaptively fuses the latent patterns exhibited in different spaces through a dynamic fusion mechanism. Experiments validate the model’s performance, achieving state-of-the-art results on multiple public datasets. Visualizations and quantitative analyses comprehensively demonstrate the ability of multi-dimensional RHR-MLPs to naturally decompose time series and reveal the tendency of higher-dimensional hypercomplex spaces to capture lower-frequency features.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting cites Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
- Are Transformers Effective for Time Series Forecasting? cites Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting
- Are Transformers Effective for Time Series Forecasting? cites Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
- iTransformer: Inverted Transformers Are Effective for Time Series Forecasting cites Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting
- iTransformer: Inverted Transformers Are Effective for Time Series Forecasting cites Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
- iTransformer: Inverted Transformers Are Effective for Time Series Forecasting cites Are Transformers Effective for Time Series Forecasting?
Builds on19
- Informer: Beyond Efficient Transformer for Long Sequence Time-Series ForecastingHaoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang et al.AAAI 2021 · 7,289 citations
- Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series ForecastingHaixu Wu, Jiehui Xu, Jianmin Wang, Mingsheng LongNeurIPS 2021 · 5,824 citations
- Are Transformers Effective for Time Series Forecasting?Ailing Zeng, Muxi Chen, Lei Zhang, Qiang XuAAAI 2023 · 3,619 citations
- Adaptive Graph Convolutional Recurrent Network for Traffic ForecastingLei Bai, Lina Yao, Can Li, Xianzhi Wang et al.NeurIPS 2020 · 2,206 citations
- iTransformer: Inverted Transformers Are Effective for Time Series ForecastingYong Liu, Tengge Hu, Haoran Zhang, Haixu Wu et al.ICLR 2024 · 1,703 citations
Related papers
- Frequency-domain MLPs are More Effective Learners in Time Series ForecastingKun Yi, Qi Zhang, Wei Fan, Shoujin Wang et al.NeurIPS 2023 · 567 citations
- TimeKAN: KAN-based Frequency Decomposition Learning Architecture for Long-term Time Series ForecastingSongtao Huang, Zhen Zhao, Can Li, Lei BaiICLR 2025
- HMformer: Unleashing Transformer's Potential for Time Series Forecasting via Hierarchical Multi-Scale ModelingRenjun Huang, Han Xiao, Bingqing Li, Baili Zhang et al.AAAI 2026
- FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series ForecastingTian Zhou, Ziqing Ma, Qingsong Wen, Xue Wang et al.ICML 2022 · 2,912 citations
- Adaptive Multi-Scale Decomposition Framework for Time Series ForecastingYifan Hu, Peiyuan Liu, Peng Zhu, Dawei Cheng et al.AAAI 2025 · 60 citations
