Pointless learning (long version) - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2017

Pointless learning (long version)

Résumé

Bayesian inversion is at the heart of probabilistic programming and more generally machine learning. Understanding inversion is made difficult by the pointful (kernel-centric) point of view usually taken in the literature. We develop a pointless (kernel-free) approach to inversion. While doing so, we revisit some foundational objects of probability theory, unravel their category-theoretical underpinnings and show how pointless Bayesian inversion sits naturally at the centre of this construction .
Fichier principal
Vignette du fichier
fossacs_cameraready_long.pdf (387.75 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01429663 , version 1 (09-01-2017)
hal-01429663 , version 2 (23-01-2017)

Identifiants

  • HAL Id : hal-01429663 , version 2

Citer

Florence Clerc, Vincent Danos, Fredrik Dahlqvist, Ilias Garnier. Pointless learning (long version). 2017. ⟨hal-01429663v2⟩

Collections

ENS-PARIS PSL
238 Consultations
494 Téléchargements

Partager

Gmail Facebook X LinkedIn More