研究人员提出，优化预处理而非扩展模型架构可以显著提高时间序列预测的准确性。他们使用 Ridge 回归作为测试平台，发现最佳回溯期是特定于序列的，并且可能与预测范围呈非单调关系。在学习到的上下文比例上进行归一化以及调整跨序列超参数共享也被证明是有益的。这些经过优化的线性模型在多个基准测试中，其性能优于先前的线性方法，甚至超越了 Transformer、MLP 和 CNN 基线。 AI
影响 表明通过适当的调整，更简单、更高效的模型可以达到最先进的性能，从而可能降低计算成本。
排序理由 学术论文，展示了关于时间序列预测方法的新研究成果。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2606.27282v1 Announce Type: new Abstract: Time-series forecasting research has been moving steadily toward larger architectures, from specialized transformers to general-purpose foundation models, on the assumption that capacity is what unlocks accuracy. We take the opposit…
Time-series forecasting research has been moving steadily toward larger architectures, from specialized transformers to general-purpose foundation models, on the assumption that capacity is what unlocks accuracy. We take the opposite position: most of the gap can be closed at far…
Time-series forecasting research has been moving steadily toward larger architectures, from specialized transformers to general-purpose foundation models, on the assumption that capacity is what unlocks accuracy. We take the opposite position: most of the gap can be closed at far…