Current browse context:
stat.ML
Change to browse by:
References & Citations
Statistics > Machine Learning
Title: Fast $ε$-free Inference of Simulation Models with Bayesian Conditional Density Estimation
(Submitted on 20 May 2016 (v1), revised 24 Oct 2016 (this version, v3), latest version 2 Apr 2018 (v4))
Abstract: Many statistical models can be simulated forwards but have intractable likelihoods. Approximate Bayesian Computation (ABC) methods are used to infer properties of these models from data. Traditionally these methods approximate the posterior over parameters by conditioning on data being inside an $\epsilon$-ball around the observed data, which is only correct in the limit $\epsilon\!\rightarrow\!0$. Monte Carlo methods can then draw samples from the approximate posterior to approximate predictions or error bars on parameters. These algorithms critically slow down as $\epsilon\!\rightarrow\!0$, and in practice draw samples from a broader distribution than the posterior. We propose a new approach to likelihood-free inference based on Bayesian conditional density estimation. Preliminary inferences based on limited simulation data are used to guide later simulations. In some cases, learning an accurate parametric representation of the entire true posterior distribution requires fewer model simulations than Monte Carlo ABC methods need to produce a single sample from an approximate posterior.
Submission history
From: George Papamakarios [view email][v1] Fri, 20 May 2016 14:34:38 GMT (2235kb,D)
[v2] Mon, 13 Jun 2016 14:55:38 GMT (2235kb,D)
[v3] Mon, 24 Oct 2016 17:11:50 GMT (2235kb,D)
[v4] Mon, 2 Apr 2018 16:05:09 GMT (2235kb,D)
Link back to: arXiv, form interface, contact.