Automatic Posterior Transformation for Likelihood-Free Inference

By David S. Greenberg, Marcel Nonnenmacher, Jakob H. Macke. International Conference on Machine Learning, 2019

457 citations recorded by the source corpus.

How can one perform Bayesian inference on stochastic simulators with intractable likelihoods? A recent approach is to learn the posterior from adaptively proposed simulations using neural network-based conditional density estimators. However, existing methods are limited to a narrow range of proposal distributions or require importance weighting that can limit performance in practice. Here we present automatic posterior transformation (APT), a new sequential neural posterior estimation method for simulation-based inference. APT can modify the posterior estimate using arbitrary, dynamically updated proposals, and is compatible with powerful flow-based density estimators. It is more flexible, scalable and efficient than previous simulation-based inference techniques. APT can operate directly on high-dimensional time series and image data, opening up new applications for likelihood-free inference.

Subfields: Artificial Intelligence, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/9d5233ea2135e7bc9023127af5b11a6af92e19cd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistical and Nonlinear Physics

Citations in the corpus

sbi: A toolkit for simulation-based inference (2020, methodology): “Currently, SNPE-C (Greenberg et al., 2019) is implemented in sbi .”
Likelihood-free MCMC with Amortized Approximate Ratio Estimators (2019, methodology): “As such, we provide this table for historic reasons to comply with previous studies such as Papamakarios & Murray (2016), Boelts et al. (2019) and Greenberg et al. (2019). a detector response x o .”
Normalizing Flows for Probabilistic Modeling and Inference (2019, methodology): “1, we can fit a flow-based model q(η |x) conditioned on x to the generated dataset {(ηn,xn)}Nn=1 in order to approximate the posterior p(η |x) (Greenberg et al., 2019; Gonçalves et al., 2019).”
Likelihood-Free Inference by Ratio Estimation (2016, methodology): “Durkan et al. (2020) discuss how inference methods in the LFIRE framework relate to conditional density LFIRE estimation methods, in particular drawing connections between the work by Hermans et al. (2020) and Greenberg et al. (2019).”
Mining gold from implicit models to improve likelihood-free inference (2018, methodology): “They can be used to train a surrogate for the likelihood p(x|θ) [17, 19, 34] or, in a Bayesian setting, the posterior p(θ|x) [29, 33, 37].”