Séminaire LATECE: par Marco Pedersoli

Bonjour à tous,

Nous avons le plaisir de vous inviter au quatrième séminaire de la session d'automne 2021 du LATECE. 

Heure et date : Le 3 novembre 2021 à 12h30

Lieu: réunion Zoom : https://uqam.zoom.us/j/88240120878

Conférencier : Marco Pedersoli

Résumé : In this talk, I present a class of techniques based on Monte Carlo Sampling that can reduce the computational cost and memory footprint of training a Deep Learning Model, while keeping the original model accuracy. As training a Deep Learning Model involves evaluating a model on the same data several times (once per training epoch), with Monte Carlo Sampling we obtain a cheap estimation of a given variable, which allows us to still optimize the same objective, while saving memory and computation. This principle is effectively used in three different computer vision applications: Emotion Recognition, Weakly Supervised Object Detection and Bayesian Object Localization.

Bio: Marco Pedersoli is Assistant Professor at ETS Montreal. He obtained his PhD in computer science in 2012 at the Autonomous University of Barcelona and the Computer Vision Center of Barcelona. Then, he was a postdoctoral fellow in computer vision and machine learning at KU Leuven with Prof. Tuytelaars and later at INRIA Grenoble with Drs. Verbeek and Schmid. At ETS Montreal he is a member of LIVIA and he is co-chairing an industrial Chair on Embedded Neural Networks for Connected Building Control. His research is mostly applied on visual recognition, the automatic interpretation and understanding of images and videos. His specific focus is on reducing the complexity and the amount of annotation required for deep learning algorithms such as convolutional and recurrent neural networks. Prof. Pedersoli has authored more than 40 publications in top-tier international conferences and journals in computer vision and machine learning. 

clockCreated with Sketch.Date / heure

mercredi 3 novembre 2021
12 h 30

pinCreated with Sketch.Lieu

UQAM - En ligne
zoom
00

personCreated with Sketch.Renseignements

Mots-clés

Groupes