Score Matched Conditional Exponential Families for Likelihood-Free Inference

12/20/2020
by   Lorenzo Pacchiardi, et al.
0

To perform Bayesian inference for stochastic simulator models for which the likelihood is not accessible, Likelihood-Free Inference (LFI) relies on simulations from the model. Standard LFI methods can be split according to how these simulations are used: to build an explicit Surrogate Likelihood, or to accept/reject parameter values according to a measure of distance from the observations (Approximate Bayesian Computation (ABC)). In both cases, simulations are adaptively tailored to the value of the observation. Here, we generate parameter-simulation pairs from the model independently on the observation, and use them to learn a conditional exponential family likelihood approximation; to parametrize it, we use Neural Networks whose weights are tuned with Score Matching. With our likelihood approximation, we can employ MCMC for doubly intractable distributions to draw samples from the posterior for any number of observations without additional model simulations, with performance competitive to comparable approaches. Further, the sufficient statistics of the exponential family can be used as summaries in ABC, outperforming the state-of-the-art method in five different models with known likelihood. Finally, we apply our method to a challenging model from meteorology.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/10/2022

Sequential Neural Score Estimation: Likelihood-Free Inference with Conditional Score Based Diffusion Models

We introduce Sequential Neural Posterior Score Estimation (SNPSE) and Se...
research
02/10/2020

On Contrastive Learning for Likelihood-free Inference

Likelihood-free methods perform parameter inference in stochastic simula...
research
11/21/2018

Sequential Neural Methods for Likelihood-free Inference

Likelihood-free inference refers to inference when a likelihood function...
research
10/23/2019

Optimistic Distributionally Robust Optimization for Nonparametric Likelihood Approximation

The likelihood function is a fundamental component in Bayesian statistic...
research
05/31/2022

Likelihood-Free Inference with Generative Neural Networks via Scoring Rule Minimization

Bayesian Likelihood-Free Inference methods yield posterior approximation...
research
04/08/2021

Generalized Bayesian Likelihood-Free Inference Using Scoring Rules Estimators

We propose a framework for Bayesian Likelihood-Free Inference (LFI) base...
research
05/30/2018

Mining gold from implicit models to improve likelihood-free inference

Simulators often provide the best description of real-world phenomena; h...

Please sign up or login with your details

Forgot password? Click here to reset