Guided sequential ABC schemes for intractable Bayesian models
Preprint, 2023
We contribute to the ABC modeller’s toolbox with novel proposal samplers that are conditional to summary statistics of the data. In a sense, the proposed parameters are “guided” to rapidly reach regions of the posterior surface that are compatible with the observed data. This speeds up the convergence of these sequential samplers, thus reducing the computational effort, while preserving the accuracy in the inference. We provide a variety of guided Gaussian and copula-based samplers for both SIS-ABC and SMC-ABC easing inference for challenging case-studies, including multimodal posteriors, highly correlated posteriors, hierarchical models with high-dimensional summary statistics (180 summaries used to infer 21 parameters) and a simulation study of cell movements (using more than 400 summaries).
Approximate Bayesian computation
sequential importance sampling
copulas
sequential Monte Carlo
simulation-based inference
Author
Umberto Picchini
Chalmers, Mathematical Sciences, Applied Mathematics and Statistics
Massimiliano Tamborrino
The University of Warwick
Statistical Inference and Stochastic Modelling of Protein Folding
Swedish Research Council (VR) (2013-5167), 2014-01-01 -- 2019-12-31.
Deep learning and likelihood-free Bayesian inference for stochastic modelling
Chalmers AI Research Centre (CHAIR), 2020-01-01 -- 2024-12-31.
Swedish Research Council (VR) (2019-03924), 2020-01-01 -- 2023-12-31.
Subject Categories
Probability Theory and Statistics