Batch simulations and uncertainty quantification in Gaussian process surrogate approximate Bayesian computation

Abstract

The computational efficiency of approximate Bayesian computation (ABC) has been improved by using surrogate models such as Gaussian processes (GP). In one such promising framework the discrepancy between the simulated and observed data is modelled with a GP which is further used to form a model-based estimator for the intractable posterior. In this article we improve this approach in several ways. We develop batch-sequential Bayesian experimental design strategies to parallellise the expensive simulations. In earlier work only sequential strategies have been used. Current surrogate-based ABC methods also do not fully account the uncertainty due to the limited budget of simulations as they output only a point estimate of the ABC posterior. We propose a numerical method to fully quantify the uncertainty in, for example, ABC posterior moments. We also provide some new analysis on the GP modelling assumptions in the resulting improved framework called Bayesian ABC and discuss its connection to Bayesian quadrature (BQ) and Bayesian optimisation (BO). Experiments with toy and real-world simulation models demonstrate advantages of the proposed techniques.

arXiv commentMinor improvements and clarifications to the text over the previous version. 20 pages, 15 figures

References in corpus (15)

• Practical Bayesian Optimization of Machine Learning Algorithms
• Predictive Entropy Search for Efficient Global Optimization of Black-box Functions
• Approximate Bayesian computation (ABC) gives exact results under the assumption of model error
• Max-value Entropy Search for Efficient Bayesian Optimization
• Fast $\epsilon$-free Inference of Simulation Models with Bayesian Conditional Density Estimation
• Approximate Bayesian computation with the Wasserstein distance
• The Parallel Knowledge Gradient Method for Batch Bayesian Optimization
• GPS-ABC: Gaussian Process Surrogate Approximate Bayesian Computation
• Automatic Posterior Transformation for Likelihood-Free Inference
• Bayesian emulator optimisation for cosmology: application to the Lyman-alpha forest
• Maximizing acquisition functions for Bayesian optimization
• Sampling for Inference in Probabilistic Models with Fast Bayesian Quadrature
• Query Efficient Posterior Estimation in Scientific Experiments via Bayesian Active Learning
• Improving Quadrature for Constrained Integrands
• Variational Bayesian Monte Carlo

Cited by in corpus (2)

• Generalized Bayesian Likelihood-Free Inference
• Bayesian Calibration and Uncertainty Quantification for a Large Nutrient Load Impact Model