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
  • [1] Akeret J(2013)Cosmohammer: cosmological parameter estimation with the MCMC hammer Astron. Comput. 2 27-39
  • [2] Seehars S(2017)Geometric MCMC for infinite-dimensional inverse problems J. Comput. Phys. 335 327-351
  • [3] Amara A(2018)Multilevel sequential Monte Carlo with dimension-independent likelihood-informed proposals SIAM/ASA J. Uncertain. Quantif. 6 762-786
  • [4] Refregier A(2008)MCMC methods for diffusion bridges Stoch. Dyn. 8 319-350
  • [5] Csillaghy A(2016)Accelerated dimension-independent adaptive Metropolis SIAM J. Sci. Comput. 38 S539-S565
  • [6] Beskos A(2016)Dimension-independent likelihood-informed MCMC J. Comput. Phys. 304 109-137
  • [7] Girolami M(2014)Likelihood-informed dimension reduction for nonlinear inverse problems Inverse Probl. 30 114015-1036
  • [8] Lan S(2016)The Bayesian formulation of EIT: analysis and algorithms Inverse Probl. Imaging 10 1007-80
  • [9] Farrell PE(2013)emcee: the MCMC hammer Publ. Astron. Soc. Pac. 125 306-2490
  • [10] Stuart AM(2010)Ensemble samplers with affine invariance Commun. Appl. Math. Comput. Sci. 5 65-489