paper-with-me

Papers

A Connection Between Learning to Reject and Bhattacharyya Divergences

2025-05-08 · Alexander Soen

Learning to reject provide a learning paradigm which allows for our models to abstain from making predictions. One way to learn the rejector is to learn an ideal marginal distribution (w.r.t. the input domain) - which characterizes a hypothetical best marginal distribution - and compares it to the true marginal distribution via a density ratio. In this paper, we consider learning a joint ideal distribution over both inputs and labels; and develop a link between rejection and thresholding different statistical divergences. We further find that when one considers a variant of the log-loss, the rejector obtained by considering the joint ideal distribution corresponds to the thresholding of the skewed Bhattacharyya divergence between class-probabilities. This is in contrast to the marginal case - that is equivalent to a typical characterization of optimal rejection, Chow's Rule - which corresponds to a thresholding of the Kullback-Leibler divergence. In general, we find that rejecting via a Bhattacharyya divergence is less aggressive than Chow's Rule.

📄 PDF Abstract BibTeX arXiv:2505.05273

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Generalizing Jensen and Bregman divergences with comparative convexity and the statistical Bhattacharyya distances with comparable means

2017-02-16 · Frank Nielsen, Richard Nock

Comparative convexity is a generalization of convexity relying on abstract notions of means. We define the Jensen divergence and the Jensen diversity from the viewpoint of comparative convexity, and show how to obtain th…

Diversity

Strictly Proper Kernel Scoring Rules and Divergences with an Application to Kernel Two-Sample Hypothesis Testing

2017-04-09 · Hamed Masnadi-Shirazi

We study strictly proper scoring rules in the Reproducing Kernel Hilbert Space. We propose a general Kernel Scoring rule and associated Kernel Divergence. We consider conditions under which the Kernel Score is strictly p…

One-class classifierscoring ruleTwo-sample testing

A generalization of the Jensen divergence: The chord gap divergence

2017-09-29 · Frank Nielsen

We introduce a novel family of distances, called the chord gap divergences, that generalizes the Jensen divergences (also called the Burbea-Rao distances), and study its properties. It follows a generalization of the cel…

Clustering

Generalized Bhattacharyya and Chernoff upper bounds on Bayes error using quasi-arithmetic means

2014-01-20 · Frank Nielsen

Bayesian classification labels observations based on given prior information, namely class-a priori and class-conditional probabilities. Bayes' risk is the minimum expected classification cost that is achieved by the Bay…

General Classification

Cumulant-free closed-form formulas for some common (dis)similarities between densities of an exponential family

2020-03-05 · Frank Nielsen, Richard Nock

It is well-known that the Bhattacharyya, Hellinger, Kullback-Leibler, $\alpha$-divergences, and Jeffreys' divergences between densities belonging to a same exponential family have generic closed-form formulas relying on …

Form