MotivatedbynonparametricposteriorinferenceinaBayesian problem where two independent unlabelled samples are available, we study the problem of sampling exactly in the set of coagulated parti- tions. These are obtained by matching a random number of cluster pairs chosen across the samples, and listing the remaining clusters unmatched. We propose an exact method for sampling from the correct target dis- tribution, and present empirical evidence in comparison with the exact distribution.
Exact sampling of two coagulated partitions
Marco Dalla Pria
;Matteo Ruggiero;
2024-01-01
Abstract
MotivatedbynonparametricposteriorinferenceinaBayesian problem where two independent unlabelled samples are available, we study the problem of sampling exactly in the set of coagulated parti- tions. These are obtained by matching a random number of cluster pairs chosen across the samples, and listing the remaining clusters unmatched. We propose an exact method for sampling from the correct target dis- tribution, and present empirical evidence in comparison with the exact distribution.
File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
2024-SIS-coag.pdf
Accesso riservato
Tipo di file:
POSTPRINT (VERSIONE FINALE DELL’AUTORE)
Dimensione
240.22 kB
Formato
Adobe PDF
|
240.22 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
