Title :

Pseudo-Bayesian Learning with Kernel Fourier Transform as Prior

Author(s) :

Letarte, Gaël [Auteur]
Département d'informatique et de génie logiciel [Québec]
Morvant, Emilie [Auteur]
Laboratoire Hubert Curien [LabHC]
Germain, Pascal [Auteur]
MOdel for Data Analysis and Learning [MODAL]

Conference title :

The 22nd International Conference on Artificial Intelligence and Statistics

City :

Naha

Country :

Japon

Start date of the conference :

2019-04-16

Journal title :

Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS) 2019,

Publication date :

2019

English keyword(s) :

Kernel Approximation
Random Fourier Features
PAC-Bayesian Theory
Kernel Learning

HAL domain(s) :

Statistiques [stat]/Machine Learning [stat.ML]

English abstract : [en]

We revisit Rahimi and Recht (2007)’s kernel random Fourier features (RFF) method through the lens of the PAC-Bayesian theory. While the primary goal of RFF is to approximate a kernel, we look at the Fourier transform as a ...
Show more >We revisit Rahimi and Recht (2007)’s kernel random Fourier features (RFF) method through the lens of the PAC-Bayesian theory. While the primary goal of RFF is to approximate a kernel, we look at the Fourier transform as a prior distribution over trigonometric hypotheses. It naturally suggests learning a posterior on these hypotheses. We derive generalization bounds that are optimized by learning a pseudo-posterior obtained from a closed-form expression. Based on this study, we consider two learning strategies: The first one finds a compact landmarks-based representation of the data where each landmark is given by a distribution-tailored similarity measure, while the second one provides a PAC-Bayesian justification to the kernel alignment method of Sinha and Duchi (2016).Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

ANR Project :

Une Perspective PAC-Bayésienne de l'Apprentissage de Représentations

Collections :

• Laboratoire Paul Painlevé - UMR 8524

Source :

Harvested from HAL