A transport approach to sequential simulation-based inference
This method addresses parameter estimation in complex noise models with nuisance parameters and black-box forward models, but appears incremental as it builds on existing transport map techniques.
The authors tackled the problem of sequential Bayesian inference for static model parameters by developing a transport-based approach that extracts conditional distributions via structured transport maps, enabling explicit surrogate models for likelihood functions and gradients. The method was applied to ice thickness characterization with conductivity measurements, though no concrete numerical results were reported in the abstract.
We present a new transport-based approach to efficiently perform sequential Bayesian inference of static model parameters. The strategy is based on the extraction of conditional distribution from the joint distribution of parameters and data, via the estimation of structured (e.g., block triangular) transport maps. This gives explicit surrogate models for the likelihood functions and their gradients. This allow gradient-based characterizations of posterior density via transport maps in a model-free, online phase. This framework is well suited for parameter estimation in case of complex noise models including nuisance parameters and when the forward model is only known as a black box. The numerical application of this method is performed in the context of characterization of ice thickness with conductivity measurements.