arXiv：cs.LG· Mihai Bogdan Deaconu, Ioan Daniel Pop·· 4 小时前AI 评分29
HAN-Mamba：用于多尺度金融波动率预测的层次化选择性状态空间网络
HAN-Mamba: Hierarchical Selective State Space Networks for Multi-Scale Financial Volatility Forecasting
AI 导读
研究者提出 HAN-Mamba，用选择性状态空间（Mamba）编码器替换 HAN-T 中的 Transformer 注意力编码器，仅在跨尺度融合阶段保留注意力机制。
正文
Abstract:Short-horizon realized volatility forecasting requires the integration of market information that evolves at incompatible temporal resolutions, from second-level order book dynamics to weekly regime drift. Our conference work introduced HAN-T, a hierarchical architecture in which scale-specific Transformer encoders process short, mid, and long-horizon streams and a learned attention fuser weighs their contributions. This article replaces the quadratic attention encoders with selective state space (Mamba) encoders while retaining attention only in the fuser, where the input is a three-token set rather than a long sequence. The resulting hybrid, HAN-Mamba, summarizes each stream through a recurrent state whose input-dependent gating matches two structural properties of volatility: persistent but decaying memory and abrupt regime shifts. On the Optiver Realized Volatility Prediction benchmark under time-aware five-fold cross-validation, HAN-Mamba improves mean RMSPE over HAN-T (0.1942 vs. 0.1965) with 33% fewer parameters. Its linear-time encoders further allow the high-frequency context to be extended from 60 to 240 buckets, reducing error to 0.1927 where the attention variant saturates, and support constant-time streaming updates at inference. Ablations attribute the gains to the encoder swap, confirm that the hierarchical prior transfers across sequence-model families, and show that the permutation-invariant attention fuser remains the correct mechanism for cross-scale integration.
| Comments: | 16 pages. Accepted for publication in Springer Lecture Notes in Artificial Intelligence (ICAART 2026 Revised Selected Papers). Extended version of the ICAART 2026 paper (DOI: https://doi.org/10.5220/0014264900004052) |
| Subjects: | Machine Learning (cs.LG); Computational Engineering, Finance, and Science (cs.CE) |
| Cite as: | arXiv:2610.10323 [cs.LG] |
| (or arXiv:2610.10323v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.10323 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Mihai Bogdan Deaconu [view email]
[v1]
Wed, 7 Oct 2026 16:12:52 UTC (28 KB)
来源：arXiv：cs.LG · arxiv.org