Variational Bayesian Monte Carlo with Noisy Likelihoods

Abstract

Variational Bayesian Monte Carlo (VBMC) is a recently introduced framework that uses Gaussian process surrogates to perform approximate Bayesian inference in models with black-box, non-cheap likelihoods. In this work, we extend VBMC to deal with noisy log-likelihood evaluations, such as those arising from simulation-based models. We introduce new `global' acquisition functions, such as expected information gain (EIG) and variational interquantile range (VIQR), which are robust to noise and can be efficiently evaluated within the VBMC setting. In a novel, challenging, noisy-inference benchmark comprising of a variety of models with real datasets from computational and cognitive neuroscience, VBMC+VIQR achieves state-of-the-art performance in recovering the ground-truth posteriors and model evidence. In particular, our method vastly outperforms `local' acquisition functions and other surrogate-based inference methods while keeping a small algorithmic cost. Our benchmark corroborates VBMC as a general-purpose technique for sample-efficient black-box Bayesian inference also with noisy models.

arXiv commentTo appear in Advances in Neural Information Processing Systems 33 (NeurIPS 2020). 26 pages, 11 figures

References in corpus (15)

• Auto-Encoding Variational Bayes
• 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
• Fast $\epsilon$-free Inference of Simulation Models with Bayesian Conditional Density Estimation
• Flexible statistical inference for mechanistic models of neural dynamics
• Yes, but Did It Work?: Evaluating Variational Inference
• Automatic Posterior Transformation for Likelihood-Free Inference
• Variational Boosting: Iteratively Refining Posterior Approximations
• Unbiased and Efficient Log-Likelihood Estimation with Inverse Binomial Sampling
• Sampling for Inference in Probabilistic Models with Fast Bayesian Quadrature
• Frank-Wolfe Bayesian Quadrature: Probabilistic Integration with Theoretical Guarantees
• Convergence Guarantees for Adaptive Bayesian Quadrature Methods
• Variational Bayesian Monte Carlo
• Evaluating Gaussian Process Metamodels and Sequential Designs for Noisy Level Set Estimation
• Automated Model Selection with Bayesian Quadrature

Cited by in corpus (5)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• Unbiased and Efficient Log-Likelihood Estimation with Inverse Binomial Sampling
• A switching Gaussian process latent force model for the identification of mechanical systems with a discontinuous nonlinearity
• PyVBMC: Efficient Bayesian inference in Python
• A survey of Monte Carlo methods for noisy and costly densities with application to reinforcement learning and ABC