Maximum likelihood estimation for the $λ$-exponential family

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tian, Xiwei, Wong, Ting-Kam Leonard, Yang, Jiaowen, Zhang, Jun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908352000294912
author Tian, Xiwei
Wong, Ting-Kam Leonard
Yang, Jiaowen
Zhang, Jun
author_facet Tian, Xiwei
Wong, Ting-Kam Leonard
Yang, Jiaowen
Zhang, Jun
contents The $λ$-exponential family generalizes the standard exponential family via a generalized convex duality motivated by optimal transport. It is the constant-curvature analogue of the exponential family from the information-geometric point of view, but the development of computational methodologies is still in an early stage. In this paper, we propose a fixed point iteration for maximum likelihood estimation under i.i.d.~sampling, and prove using the duality that the likelihood is monotone along the iterations. We illustrate the algorithm with the $q$-Gaussian distribution and the Dirichlet perturbation.
format Preprint
id arxiv_https___arxiv_org_abs_2505_03582
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Maximum likelihood estimation for the $λ$-exponential family
Tian, Xiwei
Wong, Ting-Kam Leonard
Yang, Jiaowen
Zhang, Jun
Statistics Theory
The $λ$-exponential family generalizes the standard exponential family via a generalized convex duality motivated by optimal transport. It is the constant-curvature analogue of the exponential family from the information-geometric point of view, but the development of computational methodologies is still in an early stage. In this paper, we propose a fixed point iteration for maximum likelihood estimation under i.i.d.~sampling, and prove using the duality that the likelihood is monotone along the iterations. We illustrate the algorithm with the $q$-Gaussian distribution and the Dirichlet perturbation.
title Maximum likelihood estimation for the $λ$-exponential family
topic Statistics Theory
url https://arxiv.org/abs/2505.03582