GPS-ABC: Gaussian Process Surrogate Approximate Bayesian Computation

Abstract

Scientists often express their understanding of the world through a computationally demanding simulation program. Analyzing the posterior distribution of the parameters given observations (the inverse problem) can be extremely challenging. The Approximate Bayesian Computation (ABC) framework is the standard statistical tool to handle these likelihood free problems, but they require a very large number of simulations. In this work we develop two new ABC sampling algorithms that significantly reduce the number of simulations necessary for posterior inference. Both algorithms use confidence estimates for the accept probability in the Metropolis Hastings step to adaptively choose the number of necessary simulations. Our GPS-ABC algorithm stores the information obtained from every simulation in a Gaussian process which acts as a surrogate function for the simulated statistics. Experiments on a challenging realistic biological problem illustrate the potential of these algorithms.

References in corpus (5)

• Practical Bayesian Optimization of Machine Learning Algorithms
• A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning
• Approximate Bayesian computation scheme for parameter inference and model selection in dynamical systems
• The pseudo-marginal approach for efficient Monte Carlo computations
• Coupled MCMC with a randomized acceptance probability

Cited by in corpus (26)

• The frontier of simulation-based inference
• Bayesian Optimization for Likelihood-Free Inference of Simulator-Based Statistical Models
• Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows
• Variational Bayes with Synthetic Likelihood
• Efficient acquisition rules for model-based approximate Bayesian computation
• K2-ABC: Approximate Bayesian Computation with Kernel Embeddings
• Likelihood-free MCMC with Amortized Approximate Ratio Estimators
• A Survey of Bayesian Statistical Approaches for Big Data
• Bayesian Calibration of Force-fields from Experimental Data: TIP4P Water
• Machine Learning Accelerated Likelihood-Free Event Reconstruction in Dark Matter Direct Detection
• GP-select: Accelerating EM using adaptive subspace preselection
• Kernel Recursive ABC: Point Estimation with Intractable Likelihood
• Approximating the Likelihood in Approximate Bayesian Computation
• Gaussian Process Accelerated Feldman-Cousins Approach for Physical Parameter Inference
• Batch simulations and uncertainty quantification in Gaussian process surrogate approximate Bayesian computation
• Neural Density Estimation and Likelihood-free Inference
• Differentiable Likelihoods for Fast Inversion of 'Likelihood-Free' Dynamical Systems
• Automatic Variational ABC
• Gaussian process modeling in approximate Bayesian computation to estimate horizontal gene transfer in bacteria
• Approximate Bayesian inference from noisy likelihoods with Gaussian process emulated MCMC
• A review of Approximate Bayesian Computation methods via density estimation: inference for simulator-models
• Bayesian Deconditional Kernel Mean Embeddings
• Warped Gradient-Enhanced Gaussian Process Surrogate Models for Exponential Family Likelihoods with Intractable Normalizing Constants
• Hamiltonian Monte Carlo Acceleration Using Surrogate Functions with Random Bases
• Approximate Bayesian Computation for Physical Inverse Modeling
• Inverse Gaussian Process regression for likelihood-free inference