Sparse reconstructions from few noisy data: analysis of hierarchical Bayesian models with generalized gamma hyperpriors

Sparse reconstructions from few noisy data: analysis of hierarchical Bayesian models with generalized gamma hyperpriors
复制标题

来自少量噪声数据的稀疏重建:具有广义伽马超先验的分层贝叶斯模型分析

DOI:
10.1088/1361-6420/ab4d92
复制
发表时间:
2020
期刊:
影响因子:
2.1
通讯作者:
Strang, Alexander
Strang, Alexander
中科院分区:
数学2区
文献类型:
--
作者:
Calvetti, Daniela;Pragliola, Monica;Somersalo, Erkki;Strang, Alexander

文献摘要

参考文献

被引文献

相似文献

求解稀疏促进正则化惩罚的反问题可以在贝叶斯框架下转化为寻找具有稀疏促进先验的最大后验概率(MAP)估计。在后一种情况下,在计算上方便的先验选择是条件高斯分层模型族,对于该族,未知分量的先验方差是独立的,并且遵循来自广义伽马族的超先验。在这篇文章中,我们分析了MAP估计背后的优化问题,并确定了导致全局或局部凸优化问题的超参数组合。MAP估计问题使用计算效率高的交替迭代算法来解决。分析了它在广义Gamma超模型中的性质,以及它与一些已知的稀疏性提升惩罚方法之间的联系。算例说明了该算法的收敛和稀疏性。
Solving inverse problems with sparsity promoting regularizing penalties can be recast in the Bayesian framework as finding a maximum a posteriori (MAP) estimate with sparsity promoting priors. In the latter context, a computationally convenient choice of prior is the family of conditionally Gaussian hierarchical models for which the prior variances of the components of the unknown are independent and follow a hyperprior from a generalized gamma family. In this paper, we analyze the optimization problem behind the MAP estimation and identify hyperparameter combinations that lead to a globally or locally convex optimization problem. The MAP estimation problem is solved using a computationally efficient alternating iterative algorithm. Its properties in the context of the generalized gamma hypermodel and its connections with some known sparsity promoting penalty methods are analyzed. Computed examples elucidate the convergence and sparsity promoting properties of the algorithm.
DOI: --
发表时间: 2009-12
期刊: 2009 17th European Signal Processing Conference
影响因子: --
作者:
S. D. Babacan;L. Mancera;R. Molina;A. Katsaggelos
通讯作者: S. D. Babacan;L. Mancera;R. Molina;A. Katsaggelos
DOI: 10.1137/080723995
发表时间: 2009-01-01
影响因子: 2.1
作者:
Calvetti, Daniela;Hakula, Harri;Somersalo, Erkki
通讯作者: Somersalo, Erkki
贝叶斯反演视角下线性离散不适定问题的预处理迭代方法
DOI: --
发表时间: 2007
期刊:
影响因子: --
作者:
D. Calvetti
通讯作者: D. Calvetti
基于先验 CGLS 的准 MAP 估计、统计停止规则和先验排序
DOI: 10.1137/16m108272x
发表时间: 2017
期刊: SIAM J. Sci. Comput.
影响因子: --
作者:
D. Calvetti;F. Pitolli;J. Prezioso;E. Somersalo;B. Vantaggi
通讯作者: B. Vantaggi
DOI: 10.1073/pnas.0437847100
发表时间: 2003-03-04
影响因子: 11.1
作者:
Donoho, DL;Elad, M
通讯作者: Elad, M