Kernel Stein Discrepancy on Lie Groups: Theory and Applications

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Qu, Xiaoda, Fan, Xiran, Vemuri, Baba C.
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910608476078080
author Qu, Xiaoda
Fan, Xiran
Vemuri, Baba C.
author_facet Qu, Xiaoda
Fan, Xiran
Vemuri, Baba C.
contents Distributional approximation is a fundamental problem in machine learning with numerous applications across all fields of science and engineering and beyond. The key challenge in most approximation methods is the need to tackle the intractable normalization constant pertaining to the parametrized distributions used to model the data. In this paper, we present a novel Stein operator on Lie groups leading to a kernel Stein discrepancy (KSD) which is a normalization-free loss function. We present several theoretical results characterizing the properties of this new KSD on Lie groups and its minimizers namely, the minimum KSD estimator (MKSDE). Proof of several properties of MKSDE are presented, including strong consistency, CLT and a closed form of the MKSDE for the von Mises-Fisher distribution on SO(N). Finally, we present experimental evidence depicting advantages of minimizing KSD over maximum likelihood estimation.
format Preprint
id arxiv_https___arxiv_org_abs_2305_12551
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Kernel Stein Discrepancy on Lie Groups: Theory and Applications
Qu, Xiaoda
Fan, Xiran
Vemuri, Baba C.
Statistics Theory
Probability
62R30 (Primary) 62F03, 62F05, 62F10 (Secondary)
Distributional approximation is a fundamental problem in machine learning with numerous applications across all fields of science and engineering and beyond. The key challenge in most approximation methods is the need to tackle the intractable normalization constant pertaining to the parametrized distributions used to model the data. In this paper, we present a novel Stein operator on Lie groups leading to a kernel Stein discrepancy (KSD) which is a normalization-free loss function. We present several theoretical results characterizing the properties of this new KSD on Lie groups and its minimizers namely, the minimum KSD estimator (MKSDE). Proof of several properties of MKSDE are presented, including strong consistency, CLT and a closed form of the MKSDE for the von Mises-Fisher distribution on SO(N). Finally, we present experimental evidence depicting advantages of minimizing KSD over maximum likelihood estimation.
title Kernel Stein Discrepancy on Lie Groups: Theory and Applications
topic Statistics Theory
Probability
62R30 (Primary) 62F03, 62F05, 62F10 (Secondary)
url https://arxiv.org/abs/2305.12551