Ecole d'ingénieur et centre de recherche en Sciences du numérique

Practical and scalable probabilistic machine learning

Michiardi, Pietro

SMARTDATA 2018, Keynote Speech at 3rd Workshop for the Interdepartmental Center SmartData@PoliTO, September 24-25, 2018, Turin, Italy

In this talk I will first present an overview of the research challenges we address in my group, in the context of probabilistic machine learning. With this context in mind, I will delve into the details of recent approaches to achieve a practical and scalable way to train deep probabilistic models built on the concept of Gaussian Processes. To this end, I will first briefly introduce what Gaussian Process inference is, then I will move on to describe a model approximation technique, based on random Fourier features, and approximate variational inference to train the resulting model. The talk will conclude with a distributed systems flavour, covering methods and systems to train large scale machine learning models, including data and model parallelism approaches.

Bibtex

Titre:Practical and scalable probabilistic machine learning
Type:Talk
Langue:English
Ville:Turin
Pays:ITALIE
Date:
Département:Data Science
Eurecom ref:5677
Copyright: © EURECOM. Personal use of this material is permitted. The definitive version of this paper was published in SMARTDATA 2018, Keynote Speech at 3rd Workshop for the Interdepartmental Center SmartData@PoliTO, September 24-25, 2018, Turin, Italy and is available at :
Bibtex: @talk{EURECOM+5677, year = {2018}, title = {{P}ractical and scalable probabilistic machine learning}, author = {{M}ichiardi, {P}ietro}, number = {EURECOM+5677}, month = {09}, institution = {Eurecom} address = {{T}urin, {ITALIE}}, url = {http://www.eurecom.fr/publication/5677} }
Voir aussi: