Sequential Neural Posterior and Likelihood Approximation

Abstract

We introduce the sequential neural posterior and likelihood approximation (SNPLA) algorithm. SNPLA is a normalizing flows-based algorithm for inference in implicit models, and therefore is a simulation-based inference method that only requires simulations from a generative model. SNPLA avoids Markov chain Monte Carlo sampling and correction-steps of the parameter proposal function that are introduced in similar methods, but that can be numerically unstable or restrictive. By utilizing the reverse KL divergence, SNPLA manages to learn both the likelihood and the posterior in a sequential manner. Over four experiments, we show that SNPLA performs competitively when utilizing the same number of model simulations as used in other methods, even though the inference problem for SNPLA is more complex due to the joint learning of posterior and likelihood function. Due to utilizing normalizing flows SNPLA generates posterior draws much faster (4 orders of magnitude) than MCMC-based methods.

arXiv comment28 pages, 8 tables, 14 figures. The supplementary material is attached to the main paper

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Variational Inference with Normalizing Flows
• Normalizing Flows: An Introduction and Review of Current Methods
• The frontier of simulation-based inference
• Density estimation using Real NVP
• The pseudo-marginal approach for efficient Monte Carlo computations
• Adaptive approximate Bayesian computation
• Normalizing Flows for Probabilistic Modeling and Inference
• Validating Bayesian Inference Algorithms with Simulation-Based Calibration
• Fast $\epsilon$-free Inference of Simulation Models with Bayesian Conditional Density Estimation
• Flexible statistical inference for mechanistic models of neural dynamics
• Automatic Posterior Transformation for Likelihood-Free Inference
• On Contrastive Learning for Likelihood-free Inference
• Benchmarking Simulation-Based Inference
• Partially Exchangeable Networks and Architectures for Learning Summary Statistics in Approximate Bayesian Computation
• Sequential Neural Methods for Likelihood-free Inference

Cited by in corpus (2)

• Some models are useful, but how do we know which ones? Towards a unified Bayesian model taxonomy
• Guided sequential ABC schemes for intractable Bayesian models