Mamba4Cast用状态空间模型实现高效零样本时间序列预测。
Mamba4Cast: Efficient Zero-Shot Time Series Forecasting with State Space Models
- 基于Mamba架构，单次前向传播完成全周期预测。
- 仅用合成数据训练，零样本下在真实数据上表现媲美顶级模型。
- 推理速度远超基于Transformer的模型，长时序预测更优。
本文提出Mamba4Cast，一种用于时间序列预测的零样本基础模型。该模型基于Mamba架构，并受先验-数据拟合网络（PFNs）启发，无需针对特定数据集微调即可在多种时间序列任务中稳健泛化。其核心创新在于：在真实数据集上实现强零样本性能的同时，推理速度显著低于基于Transformer的时间序列基础模型。模型仅在合成数据上训练，可一次性生成整个预测时长的输出，远超传统自回归方法。实验表明，Mamba4Cast在多个数据集上的表现与当前最先进基础模型相当，且在预测长度增加时具备更强的可扩展性。代码开源地址：https://github.com/automl/Mamba4Cast。
原文摘要 · Abstract (English)
This paper introduces Mamba4Cast, a zero-shot foundation model for time series forecasting. Based on the Mamba architecture and inspired by Prior-data Fitted Networks (PFNs), Mamba4Cast generalizes robustly across diverse time series tasks without the need for dataset specific fine-tuning. Mamba4Cast's key innovation lies in its ability to achieve strong zero-shot performance on real-world datasets while having much lower inference times than time series foundation models based on the transformer architecture. Trained solely on synthetic data, the model generates forecasts for entire horizons in a single pass, outpacing traditional auto-regressive approaches. Our experiments show that Mamba4Cast performs competitively against other state-of-the-art foundation models in various data sets while scaling significantly better with the prediction length. The source code can be accessed at https://github.com/automl/Mamba4Cast.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。