Pointless learning (long version)

Abstract : 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 .
Document type :
Preprints, Working Papers, ...
Liste complète des métadonnées

Cited literature [12 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01429663
Contributor : Ilias Garnier <>
Submitted on : Monday, January 23, 2017 - 6:36:34 PM
Last modification on : Thursday, January 17, 2019 - 4:00:02 PM
Document(s) archivé(s) le : Monday, April 24, 2017 - 3:23:32 PM

File

fossacs_cameraready_long.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01429663, version 2

Collections

Citation

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

Share

Metrics

Record views

229

Files downloads

318