Convergence of adaptive sampling schemes - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles Annals of Statistics Year : 2007

Convergence of adaptive sampling schemes

A. Guillin
  • Function : Author
C.P. Robert
  • Function : Author

Abstract

In the design of efficient simulation algorithms, one is often beset with a poor choice of proposal distributions. Although the performances of a given kernel can clarify how adequate it is for the problem at hand, a permanent on-line modification of kernels causes concerns about the validity of the resulting algorithm. While the issue is quite complex and most often intractable for MCMC algorithms, the equivalent version for importance sampling algorithms can be validated quite precisely. We derive sufficient convergence conditions for a wide class of population Monte Carlo algorithms and show that Rao-Blackwellized versions asymptotically achieve an optimum in terms of a Kullback divergence criterion, while more rudimentary versions simply do not benefit from repeated updating.
Fichier principal
Vignette du fichier
RR-5485.pdf (4.55 Mo) Télécharger le fichier

Dates and versions

inria-00070522 , version 1 (19-05-2006)

Identifiers

  • HAL Id : inria-00070522 , version 1

Cite

R. Douc, A. Guillin, Jean-Michel Marin, C.P. Robert. Convergence of adaptive sampling schemes. Annals of Statistics, 2007, 35 (1), pp.420-448. ⟨inria-00070522⟩
142 View
200 Download

Share

Gmail Facebook Twitter LinkedIn More