Approximate Bayesian computational methods

By Jean-Michel Marin, Pierre Pudlo, Christian P. Robert, Robin J. Ryder. Statistics and computing, 2011

915 citations recorded by the source corpus.

Approximate Bayesian Computation (ABC) methods, also known as likelihood-free techniques, have appeared in the past ten years as the most satisfactory approach to intractable likelihood problems, first in genetics then in a broader spectrum of applications. However, these methods suffer to some degree from calibration difficulties that make them rather volatile in their implementation and thus render them suspicious to the users of more traditional Monte Carlo methods. In this survey, we study the various improvements and extensions brought on the original ABC algorithm in recent years.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/3ac96506e1ccef100166ac412c7e91de05f346fd

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 Statistics and Probability

Citations in the corpus

Warm dark matter chills out: constraints on the halo mass function and the free-streaming length of dark matter with eight quadruple-image strong gravitational lenses (2019, methodology): “This method is an implementation of a rejection algorithm in Approximate Bayesian Computing (Rubin 1984; Marin et al. 2011; Lintusaari et al. 2017), a technique applied to problems where it is possible to generate simulated data from the model, but difficult to compute the likelihood (see also…”
Likelihood-Free Inference by Ratio Estimation (2016, methodology): “For a comprehensive introduction to the ﬁeld, we refer the reader to the review papers by Beaumont (2010); Hartig et al. (2011); Marin et al. (2012); Lintusaari et al. (2017); Sisson et al. (2018).”
GPS-ABC: Gaussian Process Surrogate Approximate Bayesian Computation (2014, methodology): “The current state-of-the-art for simulation-based model inference is likelihood-free or approximate Bayesian computation methods [23, 15].”
Dynamic Likelihood-free Inference via Ratio Estimation (DIRE) (2018, methodology): “endent standard normal variables. The parameters of interest are = ( 1; 2). We used as prior the uniform distribution on the triangle dened by 1 2[ 2;2], 1 + 2 > 1, and 1 2 <1 as Marin et al. (2012). We can again generate time-series data x = (x(1);:::;x(100)) from the model and the exac”
Adversarial Variational Optimization of Non-Differentiable Simulators (2017, methodology): “The prevalence and significance of this problem has motivated an active research effort in so-called likelihood-free inference algorithms such as Approximate Bayesian Computation (ABC) and density estimation-by-comparison algorithms (Beaumont et al., 2002; Marjoram et al., 2003; Sisson et al., 2007;”
The Rate of Convergence for Approximate Bayesian Computation (2013, background): “A survey of recent developments is given by Marin et al. (2012).”
Robust generalised Bayesian inference for intractable likelihoods (2021, neutral): “…as approximate Bayesian computation and Bayesian synthetic likelihood ; Beaumont et al., 2002; Cherief-Abdellatif & Alquier, 2020; Frazier, 2020; Marin et al., 2012; Price et al., 2018; Tavaré et al., 1997), (2) auxiliary variable MCMC (such as the exchange algorithm and pseudo-marginal MCMC;…”
Under Review as a Conference Paper at Iclr 2017 Learning in Implicit Generative Models (2017, methodology): “This piques our interest in more general approaches for non-maximum likelihood and likelihood-free estimation methods, of which there is much work (Lyu, 2011; Gutmann and Hyvärinen, 2012; Hall, 2005; Marin et al., 2012; Frogner et al., 2015).”
Deep and Hierarchical Implicit Models (2017, methodology): “The typical method for dealing with implicit models is ABC (Beaumont, 2010; Marin et al., 2012).”
Approximate Models and Robust Decisions (2014, methodology): “(2003), Marin et al. (2012), Minka (2001), Ratmann et al.”
Hierarchical Implicit Models and Likelihood-Free Variational Inference (2017, background): “ABC steps around the intractable likelihood by applying summary statistics to measure the closeness of simulated samples to real observations.”
Bayesian Parameter Estimation for Latent Markov Random Fields and Social Networks (2012, methodology): “1 The Single Auxiliary Variable Method The first MCMC method that targets the true posterior in the presence of the intractable normalising constant was the single auxiliary variable method (SAVM), which originated in Møller et al. (2004) and Møller et al. (2006). Here we present an alternative just”
An introduction to sampling via measure transport (2016, background): “This is a common scenario in Bayesian inference problems with intractable likelihoods [92, 52, 23].”
Bayesian Synthetic Likelihood (2017, background): “However, ABC is known to be highly sensitive to the choice of , the discrepancy function and to a lesser extent the kernel weighting function (Marin et al. 2012).”
Implicit Causal Models for Genome-wide Association Studies (2017, neutral): “Thus we require likelihood-free methods, which assume that one can only sample from the model’s likelihood (Marin et al., 2012; Tran et al., 2017).”
Likelihood-free MCMC with Amortized Approximate Ratio Estimators (2019, methodology): “Algorithms such as ABC (Tavaré et al., 1997; Pritchard et al., 1999; Beaumont et al., 2002; Marin et al., 2011) tackle the problem of Bayesian inference by collecting proposal states θ ∼ p ( θ ) whenever an observation x produced by the forward model x ∼ p ( x | θ ) resembles an observation x o .”
A simple approach to maximum intractable likelihood estimation (2013, background): “This is perhaps an unsurprising result and seems close in spirit to the result of Marin et al. (2013) in the model selection context, although their argument is quite different, but it does have implications for implementation of AMLE in settings with large amounts of data for which the summary stat”