用语义嵌入和自适应样条编码提升时序预测精度与可解释性
ss-Mamba: Semantic-Spline Selective State-Space Model
- 融合语义嵌入与样条编码的选通状态空间模型
- 在多个数据集上超越传统Transformer，保持线性计算复杂度
- 适合需要高效、可解释时序建模的研究与工业场景
我们提出ss-Mamba，一种新型基础模型，通过在选通状态空间建模框架中整合语义感知嵌入与自适应样条时间编码，增强时间序列预测能力。基于Transformer的成功，ss-Mamba采用Mamba选通状态空间模型作为高效替代方案，在实现相当性能的同时，将计算复杂度从二次方降至线性。预训练语言模型初始化的语义索引嵌入，使模型能通过有意义的语义先验对未见序列进行有效泛化。此外，基于样条的柯尔莫戈洛夫-阿诺德网络（KAN）动态且可解释地捕捉复杂季节性和非平稳时间效应，显著优于传统时间特征编码。大量实验验证表明，ss-Mamba在准确性、鲁棒性和可解释性方面表现卓越，展现出作为传统Transformer模型在时序预测中的通用且高效的替代方案的能力。
原文摘要 · Abstract (English)
We propose ss-Mamba, a novel foundation model that enhances time series forecasting by integrating semantic-aware embeddings and adaptive spline-based temporal encoding within a selective state-space modeling framework. Building upon the recent success of Transformer architectures, ss-Mamba adopts the Mamba selective state space model as an efficient alternative that achieves comparable performance while significantly reducing computational complexity from quadratic to linear time. Semantic index embeddings, initialized from pretrained language models, allow effective generalization to previously unseen series through meaningful semantic priors. Additionally, spline-based Kolmogorov-Arnold Networks (KAN) dynamically and interpretably capture complex seasonalities and non-stationary temporal effects, providing a powerful enhancement over conventional temporal feature encodings. Extensive experimental evaluations confirm that ss-Mamba delivers superior accuracy, robustness, and interpretability, demonstrating its capability as a versatile and computationally efficient alternative to traditional Transformer-based models in time-series forecasting.
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。