• xmlui.mirage2.page-structure.header.title
    • français
    • English
  • Help
  • Login
  • Language 
    • Français
    • English
View Item 
  •   BIRD Home
  • CEREMADE (UMR CNRS 7534)
  • CEREMADE : Publications
  • View Item
  •   BIRD Home
  • CEREMADE (UMR CNRS 7534)
  • CEREMADE : Publications
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Browse

BIRDResearch centres & CollectionsBy Issue DateAuthorsTitlesTypeThis CollectionBy Issue DateAuthorsTitlesType

My Account

LoginRegister

Statistics

Most Popular ItemsStatistics by CountryMost Popular Authors
Thumbnail - No thumbnail

Reliable ABC model choice via random forests

Pudlo, Pierre; Marin, Jean-Michel; Estoup, Arnaud; Cornuet, Jean-Marie; Gautier, Mathieu; Robert, Christian P. (2016), Reliable ABC model choice via random forests, Bioinformatics, 32, 6, p. 859-866. 10.1093/bioinformatics/btv684

Type
Article accepté pour publication ou publié
External document link
https://arxiv.org/abs/1406.6288v3
Date
2016
Journal name
Bioinformatics
Volume
32
Number
6
Publisher
Oxford University Press
Published in
Paris
Pages
859-866
Publication identifier
10.1093/bioinformatics/btv684
Metadata
Show full item record
Author(s)
Pudlo, Pierre

Marin, Jean-Michel cc

Estoup, Arnaud cc

Cornuet, Jean-Marie

Gautier, Mathieu cc

Robert, Christian P.
Abstract (EN)
Motivation: Approximate Bayesian computation (ABC) methods provide an elaborate approach to Bayesian inference on complex models, including model choice. Both theoretical arguments and simulation experiments indicate, however, that model posterior probabilities may be poorly evaluated by standard ABC techniques.Results: We propose a novel approach based on a machine learning tool named random forests (RF) to conduct selection among the highly complex models covered by ABC algorithms. We thus modify the way Bayesian model selection is both understood and operated, in that we rephrase the inferential goal as a classification problem, first predicting the model that best fits the data with RF and postponing the approximation of the posterior probability of the selected model for a second stage also relying on RF. Compared with earlier implementations of ABC model choice, the ABC RF approach offers several potential improvements: (i) it often has a larger discriminative power among the competing models, (ii) it is more robust against the number and choice of statistics summarizing the data, (iii) the computing effort is drastically reduced (with a gain in computation efficiency of at least 50) and (iv) it includes an approximation of the posterior probability of the selected model. The call to RF will undoubtedly extend the range of size of datasets and complexity of models that ABC can handle. We illustrate the power of this novel methodology by analyzing controlled experiments as well as genuine population genetics datasets.Availability and implementation: The proposed methodology is implemented in the R package abcrf available on the CRAN.Contact: jean-michel.marin@umontpellier.frSupplementary information: Supplementary data are available at Bioinformatics online.
Subjects / Keywords
summary statistics; Bayesian model choice; bagging; k -nearest neighbors; Harlequin ladybird; likelihood-free methods; posterior predictive,; random forests; sparsity; error rate; subsampling; bootstrap; model selection; Approximate Bayesian computation
JEL
C52 - Model Evaluation, Validation, and Selection
C11 - Bayesian Analysis: General

Related items

Showing items related by title and author.

  • Thumbnail
    ABC random forests for Bayesian parameter inference 
    Raynal, Louis; Marin, Jean-Michel; Pudlo, Pierre; Ribatet, Mathieu; Robert, Christian P.; Estoup, Arnaud (2019) Article accepté pour publication ou publié
  • Thumbnail
    Estimation of demo-genetic model probabilities with Approximate Bayesian Computation using linear discriminant analysis on summary statistics. 
    Cornuet, Jean-Marie; Robert, Christian P.; Pudlo, Pierre; Guillemaud, Thomas; Marin, Jean-Michel; Lombaert, Eric; Estoup, Arnaud (2012) Article accepté pour publication ou publié
  • Thumbnail
    Infering population history with DIY ABC : a user-friendly approach to Approximate Bayesian Computation 
    Estoup, Arnaud; Marin, Jean-Michel; Robert, Christian P.; Beaumont, Mark A.; Santos, Filipe; Guillemaud, Thomas; Balding, David; Cornuet, Jean-Marie (2008-04) Article accepté pour publication ou publié
  • Thumbnail
    Likelihood-free model choice 
    Marin, Jean-Michel; Pudlo, Pierre; Estoup, Arnaud; Robert, Christian P. (2018) Chapitre d'ouvrage
  • Thumbnail
    Efficient learning in ABC algorithms 
    Obert, Christian P. R; Pudlo, Pierre; Marin, Jean-Michel; Cornuet, Jean-Marie; Sedki, Mohammed (2012) Document de travail / Working paper
Dauphine PSL Bibliothèque logo
Place du Maréchal de Lattre de Tassigny 75775 Paris Cedex 16
Phone: 01 44 05 40 94
Contact
Dauphine PSL logoEQUIS logoCreative Commons logo