Sampling from Arbitrary Functions via PSD Models - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2022

Sampling from Arbitrary Functions via PSD Models

Abstract

In many areas of applied statistics and machine learning, generating an arbitrary number of independent and identically distributed (i.i.d.) samples from a given distribution is a key task. When the distribution is known only through evaluations of the density, current methods either scale badly with the dimension or require very involved implementations. Instead, we take a two-step approach by first modeling the probability distribution and then sampling from that model. We use the recently introduced class of positive semi-definite (PSD) models, which have been shown to be efficient for approximating probability densities. We show that these models can approximate a large class of densities concisely using few evaluations, and present a simple algorithm to effectively sample from these models. We also present preliminary empirical results to illustrate our assertions.
Fichier principal
Vignette du fichier
arxiv.pdf (1.62 Mo) Télécharger le fichier
Vignette du fichier
experiment_hellinger_distance_1.pdf (197.82 Ko) Télécharger le fichier
Vignette du fichier
experiment_learning_4.pdf (826.96 Ko) Télécharger le fichier
Vignette du fichier
experiment_non_continuous.pdf (76.44 Ko) Télécharger le fichier
Vignette du fichier
tau_effect_3.pdf (243.99 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Format Figure, Image
Format Figure, Image
Format Figure, Image
Format Figure, Image
Origin Files produced by the author(s)

Dates and versions

hal-03386544 , version 1 (19-10-2021)
hal-03386544 , version 2 (27-10-2021)
hal-03386544 , version 3 (23-02-2022)

Identifiers

Cite

Ulysse Marteau-Ferey, Francis Bach, Alessandro Rudi. Sampling from Arbitrary Functions via PSD Models. AISTATS 2022 - 25th International Conference on Artificial Intelligence and Statistics, Mar 2022, Valencia (virtual), Spain. ⟨hal-03386544v3⟩
83 View
148 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More