Archive for Feynman-Kac formalism

IMS workshop [day 4]

Posted in pictures, Statistics, Travel, University life with tags , , , , , , , , , on August 31, 2018 by xi'an

While I did not repeat the mistake of yesterday morning, just as well because the sun was unbearably strong!, I managed this time to board a bus headed in the wrong direction and as a result went through several remote NUS campi! Missing the first talk of the day as a result. By Youssef Marzouk, with a connection between sequential Monte Carlo and optimal transport. Transport for sampling, that is. The following talk by Tiangang Cui was however related, with Marzouk a co-author, as it aimed at finding linear transforms towards creating Normal approximations to the target to be used as proposals in Metropolis algorithms. Which may sound like something already tried a zillion times in the MCMC literature, except that the setting was rather specific to some inverse problems, imposing a generalised Normal structure on the transform, then optimised by transport arguments. It is unclear to me [from just attending the talk] how complex this derivation is and how dimension steps in, but the produced illustrations were quite robust to an increase in dimension.

The remaining talks for the day were mostly particular, from Anthony Lee introducing a new and almost costless way of producing variance estimates in particle filters, exploiting only the ancestry of particles, to Mike Pitt discussing the correlated pseudo-marginal algorithm developed with George Deligiannidis and Arnaud Doucet. Which somewhat paradoxically managed to fight the degeneracy [i.e., the need for a number of terms increasing like the time index T] found in independent pseudo-marginal resolutions, moving down to almost log(T)… With an interesting connection to the quasi SMC approach of Mathieu and Nicolas. And Sebastian Reich also stressed the links with optimal transport in a talk about data assimilation that was way beyond my reach. The day concluded with fireworks, through a magistral lecture by Professeur Del Moral on a continuous time version of PMCMC using the Feynman-Kac terminology. Pierre did a superb job during his lecture towards leading the whole room to the conclusion.

Wang, Landau, Markov, and others…

Posted in pictures, Statistics, University life with tags , , , , , , , , , on April 11, 2012 by xi'an

On Thursday, the “Big’MC” seminar welcomes two talks (at 3pm and 4pm, resp., in IHP, Amphi Darboux):

  • Orateur :Pierre Jacob (ENSAE) et Robin Ryder (CEREMADE)
  • Titre : Some aspects of the Wang-Landau algorithm.
  • Résumé : The Wang-Landau algorithm is an adaptive MCMC algorithm which generates a Markov chain designed to move efficiently in the state space, by constantly penalizing already-visited regions. It hence falls into the class of exploratory algorithms, especially when the chosen regions correspond to different levels of density values. We explore two novel aspects of the Wang-Landau algorithm. First, we show that the algorithm reaches the so-called Flat Histogram criterion in finite time, which ensures convergence properties. Second, we examine the effect of using multiple chains, interacting through a common component. That component essentially represents the history of already-visited regions, computed on all the chains. We show numerically the benefit of using parallel chains even if a single processing unit is available, in terms of stabilization of the schedule used in the adaptation process. If time permits, we shall present an ongoing attempt to study theoretically the effect of parallelization using Feynman-Kac semigroups.
  • Références http://arxiv.org/abs/1110.4025 et http://arxiv.org/abs/1109.3829


and

  • Orateur : Nick Whiteley ( Univ. Bristol, UK)
  • Titre  : A particle method for approximating principal eigen-functions and related quantities
  • Résumé : Perron-Frobenius theory treats the existence of a positive eigen-vector associated with the principal eigen-value \lambda_{\star} of a non-negative matrix, say Q. A simple method for approximating this eigen-vector involves computing the iterate \lambda_{\star}^{-n}Q^{(n)}, for large n. In the more general case that Q is a non-negative integral kernel, an extended Perron-Frobenius theory applies, but it is typical that neither the principal eigen-function nor the iterate \lambda_{\star}^{-n}Q^{(n)} can be computed exactly. In this setting we introduce an interacting particle algorithm which yields a numerical approximation of the principal eigen-function and the associated twisted Markov kernel. Some of its theoretical properties will be discussed and applications will be outlined. In particular, the algorithm allows approximation of an optimal importance sampling method for Markov chain rare event estimation.
    Joint work with Nikolas Kantas.
  • Référence : http://arxiv.org/abs/1202.6678