arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.04259cs.LGcs.NAmath.NA
收藏

用于贝叶斯反问题的稳定无导数高斯混合变分推断

Stable Derivative Free Gaussian Mixture Variational Inference for Bayesian Inverse Problems

School of Mathematical Sciences, Nankai University, Tianjin, China(南开大学数学科学学院);Courant Institute, New York University, NY(纽约大学Courant研究所);School of Mathematical Sciences, Peking University, Beijing, China(北京大学数学科学学院);Beijing International Center for Mathematical Research, Center for Machine Learning Research, Peking University, Beijing, China(北京大学北京国际数学研究中心、机器学习研究中心)

Baojun Che, Yifan Chen, Zhenghao Huan, Daniel Zhengyu Huang, Weijie Wang

AI总结:

针对大规模贝叶斯反问题中正向模型昂贵、多模态和梯度不可用的问题,提出结合 Fisher-Rao 自然梯度与专门求积规则的 DF-GMVI,实现稳定、仿射不变的无导数高斯混合后验近似。

摘要:

本文关注仅已知到归一化常数的概率分布的近似,重点研究科学计算中大规模反问题的贝叶斯推断。在此背景下,关键挑战包括正向模型的重复评估代价高昂、多模态性,以及正向模型梯度不可获取。为解决这些问题,我们开发了一个变分推断框架,将 Fisher-Rao 自然梯度与专门的求积规则相结合,以实现高斯混合变分族的无导数更新。所得方法称为无导数高斯混合变分推断(Derivative Free Gaussian Mixture Variational Inference,DF-GMVI),可保证协方差正定性和仿射不变性,为近似复杂后验分布提供稳定且高效的框架。DF-GMVI 的有效性通过在具有挑战性的场景上开展数值实验得到证明,包括最高 100 维空间中具有多模态、无限多个模态以及弯曲模态的分布。该方法的实用性还在一个大规模应用中得到进一步验证:它成功地从正时刻的解数据中恢复了 Navier-Stokes 方程的初始条件。

英文摘要:

This paper is concerned with the approximation of probability distributions known up to normalization constants, with a focus on Bayesian inference for large-scale inverse problems in scientific computing. In this context, key challenges include costly repeated evaluations of forward models, multimodality, and inaccessible gradients for the forward model. To address them, we develop a variational inference framework that combines Fisher-Rao natural gradient with specialized quadrature rules to enable derivative free updates of Gaussian mixture variational families. The resulting method, termed Derivative Free Gaussian Mixture Variational Inference (DF-GMVI), guarantees covariance positivity and affine invariance, offering a stable and efficient framework for approximating complex posterior distributions. The effectiveness of DF-GMVI is demonstrated through numerical experiments on challenging scenarios, including distributions with multiple modes, infinitely many modes, and curved modes in spaces with up to 100 dimensions. The method's practicality is further demonstrated in a large-scale application, where it successfully recovers the initial conditions of the Navier-Stokes equations from solution data at positive times.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑