On the Theoretical Equivalence of Several Trade-Off Curves Assessing Statistical Proximity - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2021

On the Theoretical Equivalence of Several Trade-Off Curves Assessing Statistical Proximity

Résumé

The recent advent of powerful generative models has triggered the renewed development of quantitative measures to assess the proximity of two probability distributions. As the scalar Frechet inception distance remains popular, several methods have explored computing entire curves, which reveal the trade-off between the fidelity and variability of the first distribution with respect to the second one. Several of such variants have been proposed independently and while intuitively similar, their relationship has not yet been made explicit. In an effort to make the emerging picture of generative evaluation more clear, we propose a unification of four curves known respectively as: the precision-recall (PR) curve, the Lorenz curve, the receiver operating characteristic (ROC) curve and a special case of R\'enyi divergence frontiers. In addition, we discuss possible links between PR / Lorenz curves with the derivation of domain adaptation bounds.

Dates et versions

hal-03336929 , version 1 (07-09-2021)

Identifiants

Citer

Rodrigue Siry, Ryan Webster, Loïc Simon, Julien Rabin. On the Theoretical Equivalence of Several Trade-Off Curves Assessing Statistical Proximity. 2021. ⟨hal-03336929⟩
32 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More