Efficient acquisition rules for model-based approximate Bayesian computation

Abstract

Approximate Bayesian computation (ABC) is a method for Bayesian inference when the likelihood is unavailable but simulating from the model is possible. However, many ABC algorithms require a large number of simulations, which can be costly. To reduce the computational cost, Bayesian optimisation (BO) and surrogate models such as Gaussian processes have been proposed. Bayesian optimisation enables one to intelligently decide where to evaluate the model next but common BO strategies are not designed for the goal of estimating the posterior distribution. Our paper addresses this gap in the literature. We propose to compute the uncertainty in the ABC posterior density, which is due to a lack of simulations to estimate this quantity accurately, and define a loss function that measures this uncertainty. We then propose to select the next evaluation location to minimise the expected loss. Experiments show that the proposed method often produces the most accurate approximations as compared to common BO strategies.

30 pages, 10 figures

References in corpus (7)

• Practical Bayesian Optimization of Machine Learning Algorithms
• Approximate Bayesian computation scheme for parameter inference and model selection in dynamical systems
• Predictive Entropy Search for Efficient Global Optimization of Black-box Functions
• Max-value Entropy Search for Efficient Bayesian Optimization
• Slice sampling covariance hyperparameters of latent Gaussian models
• Efficient acquisition rules for model-based approximate Bayesian computation
• A comparison of emulation methods for Approximate Bayesian Computation

Cited by in corpus (8)

• General framework for cosmological dark matter bounds using $N$-body simulations
• Efficient acquisition rules for model-based approximate Bayesian computation
• Sequential Bayesian experimental design for calibration of expensive simulation models
• Bridging the reality gap in quantum devices with physics-aware machine learning
• A survey of Monte Carlo methods for noisy and costly densities with application to reinforcement learning and ABC
• Parametric scaling of power exhaust in EU-DEMO alternative divertor simulations
• Sequentially guided MCMC proposals for synthetic likelihoods and correlated synthetic likelihoods
• Uncertainty Quantification in Breakup Reactions