On two ways to use determinantal point ...
Document type :
Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès avec actes
Title :
On two ways to use determinantal point processes for Monte Carlo integration
Author(s) :
Gautier, Guillaume [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Sequential Learning [SEQUEL]
Bardenet, Remi [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Centre National de la Recherche Scientifique [CNRS]
Valko, Michal [Auteur]
Sequential Learning [SEQUEL]
DeepMind [London]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Sequential Learning [SEQUEL]
Bardenet, Remi [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Centre National de la Recherche Scientifique [CNRS]
Valko, Michal [Auteur]
Sequential Learning [SEQUEL]
DeepMind [London]
Conference title :
NEGDEPML 2019 - ICML Workshop on Negative Dependence in ML
City :
Long Beach, CA
Country :
Etats-Unis d'Amérique
Start date of the conference :
2019-06-14
HAL domain(s) :
Mathématiques [math]/Statistiques [math.ST]
Informatique [cs]/Apprentissage [cs.LG]
Informatique [cs]/Apprentissage [cs.LG]
English abstract : [en]
This paper focuses on Monte Carlo integration with determinantal point processes (DPPs) which enforce negative dependence between quadrature nodes. We survey the properties of two unbiased Monte Carlo estimators of the ...
Show more >This paper focuses on Monte Carlo integration with determinantal point processes (DPPs) which enforce negative dependence between quadrature nodes. We survey the properties of two unbiased Monte Carlo estimators of the integral of interest: a direct one proposed by Bardenet & Hardy (2016) and a less obvious 60-year-old estimator by Ermakov & Zolotukhin (1960) that actually also relies on DPPs. We provide an efficient implementation to sample exactly a particular multidimen-sional DPP called multivariate Jacobi ensemble. This let us investigate the behavior of both estima-tors on toy problems in yet unexplored regimes.Show less >
Show more >This paper focuses on Monte Carlo integration with determinantal point processes (DPPs) which enforce negative dependence between quadrature nodes. We survey the properties of two unbiased Monte Carlo estimators of the integral of interest: a direct one proposed by Bardenet & Hardy (2016) and a less obvious 60-year-old estimator by Ermakov & Zolotukhin (1960) that actually also relies on DPPs. We provide an efficient implementation to sample exactly a particular multidimen-sional DPP called multivariate Jacobi ensemble. This let us investigate the behavior of both estima-tors on toy problems in yet unexplored regimes.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Collections :
Source :
Files
- https://hal.archives-ouvertes.fr/hal-02160382/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02160382/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02160382/document
- Open access
- Access the document
- document
- Open access
- Access the document
- icml_camera_ready.pdf
- Open access
- Access the document
- document
- Open access
- Access the document
- icml_camera_ready.pdf
- Open access
- Access the document