Saved in:
Bibliographic Details
Main Authors: Sire, Charlie, Richet, Yann, Riche, Rodolphe Le, Rullière, Didier, Rohmer, Jérémy, Pheulpin, Lucie
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2308.10871
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910614933209088
author Sire, Charlie
Richet, Yann
Riche, Rodolphe Le
Rullière, Didier
Rohmer, Jérémy
Pheulpin, Lucie
author_facet Sire, Charlie
Richet, Yann
Riche, Rodolphe Le
Rullière, Didier
Rohmer, Jérémy
Pheulpin, Lucie
contents Quantization summarizes continuous distributions by calculating a discrete approximation. Among the widely adopted methods for data quantization is Lloyd's algorithm, which partitions the space into Voronoï cells, that can be seen as clusters, and constructs a discrete distribution based on their centroids and probabilistic masses. Lloyd's algorithm estimates the optimal centroids in a minimal expected distance sense, but this approach poses significant challenges in scenarios where data evaluation is costly, and relates to rare events. Then, the single cluster associated to no event takes the majority of the probability mass. In this context, a metamodel is required and adapted sampling methods are necessary to increase the precision of the computations on the rare clusters.
format Preprint
id arxiv_https___arxiv_org_abs_2308_10871
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle FunQuant: A R package to perform quantization in the context of rare events and time-consuming simulations
Sire, Charlie
Richet, Yann
Riche, Rodolphe Le
Rullière, Didier
Rohmer, Jérémy
Pheulpin, Lucie
Computation
Machine Learning
Quantization summarizes continuous distributions by calculating a discrete approximation. Among the widely adopted methods for data quantization is Lloyd's algorithm, which partitions the space into Voronoï cells, that can be seen as clusters, and constructs a discrete distribution based on their centroids and probabilistic masses. Lloyd's algorithm estimates the optimal centroids in a minimal expected distance sense, but this approach poses significant challenges in scenarios where data evaluation is costly, and relates to rare events. Then, the single cluster associated to no event takes the majority of the probability mass. In this context, a metamodel is required and adapted sampling methods are necessary to increase the precision of the computations on the rare clusters.
title FunQuant: A R package to perform quantization in the context of rare events and time-consuming simulations
topic Computation
Machine Learning
url https://arxiv.org/abs/2308.10871