Ensemble sampler for infinite-dimensional inverse problems

被引:0
作者
Jeremie Coullon
Robert J. Webber
机构
[1] Lancaster University,
[2] New York University,undefined
来源
Statistics and Computing | 2021年 / 31卷
关键词
Bayesian inverse problems; Markov chain Monte Carlo; Infinite-dimensional inverse problems; Dimensionality reduction; 65C05; 35R30; 62F15;
D O I
暂无
中图分类号
学科分类号
摘要
We introduce a new Markov chain Monte Carlo (MCMC) sampler for infinite-dimensional inverse problems. Our new sampler is based on the affine invariant ensemble sampler, which uses interacting walkers to adapt to the covariance structure of the target distribution. We extend this ensemble sampler for the first time to infinite-dimensional function spaces, yielding a highly efficient gradient-free MCMC algorithm. Because our new ensemble sampler does not require gradients or posterior covariance estimates, it is simple to implement and broadly applicable.
引用
收藏
相关论文
共 60 条
[11]  
Beskos A(2014)Spectral gaps for a metropolis-hastings algorithm in infinite dimensions Ann. Appl. Probab. 24 2455-138
[12]  
Jasra A(2014)Sequential Monte Carlo methods for high-dimensional inverse problems: a case study for the Navier–Stokes equations SIAM/ASA J. Uncertain. Quantif. 2 464-1092
[13]  
Law K(2014)Proposals which speed up function-space MCMC J. Comput. Appl. Math. 262 127-367
[14]  
Marzouk Y(1953)Equation of state calculations by fast computing machines J. Chem. Phys. 21 1087-343
[15]  
Zhou Y(2009)Examples of adaptive MCMC J. Comput. Graph. Stat. 18 349-639
[16]  
Beskos A(2018)On a generalization of the preconditioned Crank–Nicolson Metropolis algorithm Found. Comput. Math. 18 309-undefined
[17]  
Roberts G(2010)Inverse problems: a Bayesian perspective Acta Numer. 19 451-undefined
[18]  
Stuart A(2017)A hybrid adaptive MCMC algorithm in function spaces SIAM/ASA J. Uncertain. Quantifi. 5 621-undefined
[19]  
Voss J(undefined)undefined undefined undefined undefined-undefined
[20]  
Chen Y(undefined)undefined undefined undefined undefined-undefined