Bayesian Inversion

I began working on Bayesian inversion during my postdoctoral research at the University of Warwick, under the supervision of Andrew Stuart. In [1], we compared a range of approximate and heuristic Bayesian inversion methods and benchmarked their performance against preconditioned Crank–Nicolson (pCN) MCMC, with applications to oil reservoir models.

In subsequent work, I explored Sequential Monte Carlo methods for Bayesian inversion in the context of resin infusion into fibre reinforcements [2], motivated by applications in composite manufacturing. I have also investigated transform particle filters and related ensemble-based particle methods in [3], with a focus on scalable inference in high-dimensional settings.

References

[1]

M. A. Iglesias, K. J. H. Law, and A. M. Stuart, “Evaluation of Gaussian approximations for data assimilation in reservoir models,” Computational Geosciences, vol. 17, no. 5, pp. 851–885, 2013, doi: 10.1007/s10596-013-9359-x.

[2]

M. Iglesias, M. Park, and M. V. Tretyakov, “Bayesian inversion in resin transfer molding,” Inverse Problems, vol. 34, no. 10, p. 105002, Jul. 2018, doi: 10.1088/1361-6420/aad1cc.

[3]

S. Ruchi, S. Dubinkina, and M. A. Iglesias, “Transform-based particle filtering for elliptic bayesian inverse problems,” Inverse Problems, vol. 35, no. 11, p. 115005, Oct. 2019, doi: 10.1088/1361-6420/ab30f3.