Nos partenaires

CNRS

Rechercher





Accueil du site > Français > Evénements > Séminaires

Séminaires

 

L’IRIT étant localisé sur plusieurs sites, ses séminaires sont organisés et ont lieu soit à l’Université Toulouse 3 Paul Sabatier (UT3), l’Université Toulouse 1 Capitole (UT1), l’INP-ENSEEIHT ou l’Université Toulouse 2 Jean Jaurès (UT2J).

 

From perceptrons to Alpha Go and beyond: An overview of Deep Learning

Dennis WILSON - Equipe VORTEX - IRIT (France)

Mercredi 29 Novembre 2017, 14h00 - 16h00
UT3 Paul Sabatier, IRIT, Salle du Conseil
Version PDF :

Abstract

Deep learning has dominated the fields of machine learning and artificial intelligence recently, and has found application in everything from self-driving cars to online advertisement. In this talk, the foundations of deep learning will be briefly explored to allow for an understanding of the state of the art. Contemporary models, deep convolutional networks and recurrent networks, and training methods, from single-layer back-propagation to adversarial training, will be shown. Finally, applications from various fields will be used to showcase the utility of deep learning and the challenges of applying it, both of a technical and ethical nature.

Short Bio: Dennis G. Wilson is a PhD candidate at IRIT in the VORTEX team, studying online learning in neural networks. Previously, he worked in the Anyscale Learning for All group at MIT, applying evolutionary algorithms to the problem of wind farm layout optimization.

 

Retour