Essec\Faculty\Model\Contribution {#6196`
#_index: "academ_contributions"
#_id: "10698"
#_source: array:26 [``
"id" => "10698"
"slug" => "controlled-sequential-monte-carlo"
"yearMonth" => "2019-05"
"year" => "2019"
"title" => "Controlled Sequential Monte Carlo"
"description" => "HENG, J., BISHOP, A.N., DELIGIANNIDIS, G. et DOUCET, A. (2019). Controlled Sequential Monte Carlo. <i>Annals of Statistics</i>, 48(5), pp. 2904-2929."
"authors" => array:4 [``
0 => array:3 [``
"name" => "HENG Jeremy"
"bid" => "B00760223"
"slug" => "heng-jeremy"
`]
1 => array:1 [`
"name" => "BISHOP Adrian N."
`]
2 => array:1 [`
"name" => "DELIGIANNIDIS George"
`]
3 => array:1 [`
"name" => "DOUCET Arnaud"
`]
]
"ouvrage" => ""
"keywords" => array:6 [`
0 => "State space models"
1 => "annealed importance sampling"
2 => "normalizing constants"
3 => "optimal control"
4 => "approximate dynamic programming"
5 => "reinforcement learning"
`]
"updatedAt" => "2023-07-10 17:25:23"
"publicationUrl" => "https://projecteuclid.org/euclid.aos/1600480936"
"publicationInfo" => array:3 [`
"pages" => "2904-2929"
"volume" => "48"
"number" => "5"
`]
"type" => array:2 [`
"fr" => "Articles"
"en" => "Journal articles"
`]
"support_type" => array:2 [`
"fr" => "Revue scientifique"
"en" => "Scientific journal"
`]
"countries" => array:2 [`
"fr" => null
"en" => null
`]
"abstract" => array:2 [`
"fr" => "Sequential Monte Carlo methods, also known as particle methods, are a popular set of techniques for approximating high-dimensional probability distributions and their normalizing constants. These methods have found numerous applications in statistics and related fields; for example, for inference in nonlinear non-Gaussian state space models, and in complex static models. Like many Monte Carlo sampling schemes, they rely on proposal distributions which crucially impact their performance. We introduce here a class of controlled sequential Monte Carlo algorithms, where the proposal distributions are determined by approximating the solution to an associated optimal control problem using an iterative scheme. This method builds upon a number of existing algorithms in econometrics, physics and statistics for inference in state space models, and generalizes these methods so as to accommodate complex static models. We provide a theoretical analysis concerning the fluctuation and stability of this methodology that also provides insight into the properties of related algorithms. We demonstrate significant gains over state-of-the-art methods at a fixed computational complexity on a variety of applications."
"en" => "Sequential Monte Carlo methods, also known as particle methods, are a popular set of techniques for approximating high-dimensional probability distributions and their normalizing constants. These methods have found numerous applications in statistics and related fields; for example, for inference in nonlinear non-Gaussian state space models, and in complex static models. Like many Monte Carlo sampling schemes, they rely on proposal distributions which crucially impact their performance. We introduce here a class of controlled sequential Monte Carlo algorithms, where the proposal distributions are determined by approximating the solution to an associated optimal control problem using an iterative scheme. This method builds upon a number of existing algorithms in econometrics, physics and statistics for inference in state space models, and generalizes these methods so as to accommodate complex static models. We provide a theoretical analysis concerning the fluctuation and stability of this methodology that also provides insight into the properties of related algorithms. We demonstrate significant gains over state-of-the-art methods at a fixed computational complexity on a variety of applications."
`]
"authors_fields" => array:2 [`
"fr" => "Systèmes d’Information, Sciences de la Décision et Statistiques"
"en" => "Information Systems, Decision Sciences and Statistics"
`]
"indexedAt" => "2024-02-27T06:21:46.000Z"
"docTitle" => "Controlled Sequential Monte Carlo"
"docSurtitle" => "Articles"
"authorNames" => "<a href="/cv/heng-jeremy">HENG Jeremy</a>, BISHOP Adrian N., DELIGIANNIDIS George, DOUCET Arnaud"
"docDescription" => "<span class="document-property-authors">HENG Jeremy, BISHOP Adrian N., DELIGIANNIDIS George, DOUCET Arnaud</span><br><span class="document-property-authors_fields">Systèmes d’Information, Sciences de la Décision et Statistiques</span> | <span class="document-property-year">2019</span>"
"keywordList" => "<a href="#">State space models</a>, <a href="#">annealed importance sampling</a>, <a href="#">normalizing constants</a>, <a href="#">optimal control</a>, <a href="#">approximate dynamic programming</a>, <a href="#">reinforcement learning</a>"
"docPreview" => "<b>Controlled Sequential Monte Carlo</b><br><span>2019-05 | Articles </span>"
"docType" => "research"
"publicationLink" => "<a href="https://projecteuclid.org/euclid.aos/1600480936" target="_blank">Controlled Sequential Monte Carlo</a>"
]
+lang: "fr"
+"_type": "_doc"
+"_score": 8.728815
+"parent": null
}