Abstract:Time series forecasting (TSF) is critical in domains like energy, finance, healthcare, and logistics, requiring models that generalize across diverse datasets. Large pre-trained models such as Chronos and Time-MoE show strong zero-shot (ZS) performance but suffer from high computational costs. In this work, we introduce Super-Linear, a lightweight and scalable mixture-of-experts (MoE) model for general forecasting. It replaces deep architectures with simple frequency-specialized linear experts, trained on resampled data across multiple frequency regimes. A lightweight spectral gating mechanism dynamically selects relevant experts, enabling efficient, accurate forecasting. Despite its simplicity, Super-Linear demonstrates strong performance across benchmarks, while substantially improving efficiency, robustness to sampling rates, and interpretability. The implementation of Super-Linear is available at: \href{this https URL}{this https URL}.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2509.15105 [cs.LG] |
| (or arXiv:2509.15105v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2509.15105 arXiv-issued DOI via DataCite |
|
| Journal reference: | Transactions on Machine Learning Research (TMLR), 2026 |
From: Liran Nochumsohn [view email]
[v1]
Thu, 18 Sep 2025 16:11:31 UTC (4,473 KB)
[v2]
Tue, 27 Jan 2026 12:31:03 UTC (5,376 KB)
[v3]
Fri, 22 May 2026 12:07:12 UTC (5,516 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。