用吠陀数学思想改进时间序列建模，提升长程依赖捕捉能力。
Naga: Vedic Encoding for Deep State Space Models
- 双向处理正反序列，通过哈达玛积融合表示
- 在6个数据集上超越28个顶尖模型，效率更高
- 适合需要高效长序列建模的场景
本文提出Naga，一种受吠陀数学结构启发的深度状态空间模型编码方法。该方法通过联合处理正向与逆向时间序列，生成双向表示，并采用逐元素（哈达玛）交互方式融合，形成具有吠陀特征的编码，增强模型对远距离时间依赖的捕捉能力。我们在多个长期时间序列预测基准上进行评估，包括ETTh1、ETTh2、ETTm1、ETTm2、Weather、Traffic和ILI。实验结果表明，Naga优于28个当前最先进的模型，且相比现有深度SSM方法更具计算效率。研究显示，引入结构化、吠陀启发的分解可为长程序列建模提供可解释且高效的替代方案。
原文摘要 · Abstract (English)
This paper presents Naga, a deep State Space Model (SSM) encoding approach inspired by structural concepts from Vedic mathematics. The proposed method introduces a bidirectional representation for time series by jointly processing forward and time-reversed input sequences. These representations are then combined through an element-wise (Hadamard) interaction, resulting in a Vedic-inspired encoding that enhances the model's ability to capture temporal dependencies across distant time steps. We evaluate Naga on multiple long-term time series forecasting (LTSF) benchmarks, including ETTh1, ETTh2, ETTm1, ETTm2, Weather, Traffic, and ILI. The experimental results show that Naga outperforms 28 current state of the art models and demonstrates improved efficiency compared to existing deep SSM-based approaches. The findings suggest that incorporating structured, Vedic-inspired decomposition can provide an interpretable and computationally efficient alternative for long-range sequence modeling.
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。