arXiv：cs.LG· Paul F. V. Wiemann, Matthias Katzfuss·· 13 小时前
面向非高斯空间场的可扩展稀疏自回归变分推理
Scalable Sparse Autoregressive Variational Inference for Non-Gaussian Spatial Fields
AI 导读
研究者提出一种可扩展变分推理框架，用于非高斯空间场的潜变量场后验估计。该方法在变分族中复刻先验的稀疏最近邻自回归结构，采用归一化流条件分布提升表达能力，并通过单一共享神经调节器摊销全部变分参数。配合缩减祖先集的双重随机优化，每次迭代达到多对数复杂度。
正文
Abstract:Latent field models are essential for spatial statistics, but the posterior distribution of the latent field, even with a Gaussian process (GP) prior, is analytically intractable whenever the likelihood is non-Gaussian. Existing approximations are either too slow or too restrictive for large non-Gaussian datasets. We present a scalable variational inference framework that mirrors the sparse nearest-neighbor autoregressive structure of the prior in the variational family, uses normalizing-flow conditionals for expressiveness, and amortizes all variational parameters via a single shared neural conditioner. Doubly stochastic optimization with reduced ancestor sets achieves polylogarithmic complexity per iteration.
| Comments: | Short paper for the Joint Meeting SIS-FENStatS 2026, Rome, June 22-25, 2026 |
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG); Computation (stat.CO); Methodology (stat.ME) |
| Cite as: | arXiv:2610.13656 [stat.ML] |
| (or arXiv:2610.13656v1 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2610.13656 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Paul Wiemann [view email]
[v1]
Fri, 9 Oct 2026 14:30:12 UTC (558 KB)
来源：arXiv：cs.LG · arxiv.org