IVADO/MILA Deep Learning School 5th edition

December 2-6, 2019 | Vancouver (English)

About

Summary

Deep learning is a machine learning technique that has significantly improved previous results in computer vision, speech recognition, machine translation and other areas. Many other areas are affected by this new technology, or will be. In response to the interest generated by this technology, and in response to training needs, IVADO and Mila are partnering with the UBC Data Science Institute to offer this training in Vancouver from December 2 to 6, 2019.

Target audience

The content of this school is mainly aimed at industry professionals and SMEs with basic knowledge of mathematics and programming (engineers, computer scientists, statisticians, technical project managers, product managers, systems engineers, etc.), but professors and graduate students in science or engineering (mainly those who are not yet familiar with deep learning) may also find it interesting.

Prerequisites

A minimal knowledge of programming (ideally Python) and basic knowledge in mathematics (linear algebra, statistics) is desirable.

Objectives

At the end of the training week, participants should be able to:

  • Understand the basics and terminology related to deep learning
  • Understand the methodology for carrying out a project in deep learning
  • Identify the types of neural networks to use to solve different types of problems
  • Get familiar with deep learning libraries through practical and tutorial sessions

Dates and place

December 2-6, 2019, Vancouver
Deep Learning School (Vancouver 2019) Scholarship Application Form

UBC Nest South Ballroom
6133 University Boulevard
Vancouver, BC

View Map

Contact

For any comments, do not hesitate to contact us at the following email address: formations@ivado.ca.

Dates and places

December 2-6, 2019, Vancouver, English

Program

Machine Learning

09:00-09:10 : Welcome

Welcome words by IVADO and Mila

09:10-10:15 : Presentation

Machine learning and experimental protocol

Gaétan Marceau Caron

10:15 – 10:45 : Break

10:45 – 12:00 : Presentation

Introduction to Machine Learning

Gaétan Marceau Caron

12:00 – 13:30 : Lunch

13:30 – 14:45 : Presentation

Machine learning tools

Jeremy Pinto

14:45 – 15:15 : Break

15:15 – 16:30 : Tutorial

Data & metrics with pyTorch

Access to the public github repo for the tutorials:  https://github.com/mila-iqia/ivado-mila-dl-school-2019

Deep Learning

09:00 – 10:15 : Presentation

Introduction to deep learning

Gaétan Marceau Caron

10:15 – 10:45 : Break

10:45 – 12:00 : Presentation

Computational graph & backpropagation

Gaétan Marceau Caron

12:00 – 13:30 : Lunch

13:30 – 14:45 : Presentation

Optimization in deep learning

Gaétan Marceau Caron

14:45 – 15:15 : Break

15:15 – 16:30 : Tutorial

Categorical data with multilayer perceptron (MLP)

Access to the public github repo for the tutorials:  https://github.com/mila-iqia/ivado-mila-dl-school-2019

CNN

09:00 – 10:15 : Presentation

Introduction to convolutional neural networks, part I

Jeremy Pinto

10:15 – 10:45 : Break

10:45 – 12:00 : Presentation

Introduction to convolutional neural networks, part II

Jeremy Pinto

12:00 – 13:30 : Lunch

13:30 – 14:45 : Presentation

Convolutional neural network architectures

Jeremy Pinto

14:45 – 15:15 : Break

15:15 – 16:30 : Tutorial

Getting started with convolutional neural networks

Access to the public github repo for the tutorials:  https://github.com/mila-iqia/ivado-mila-dl-school-2019

RNN

09:00 – 10:15 : Presentation

Introduction to recurrent neural networks

Mirko Bronzi

10:15 – 10:45 : Break

10:45 – 12:00 : Presentation

Sequence to sequence models

Mirko Bronzi

12:00 – 13:30 : Lunch

13:30 – 14:45 : Presentation

Natural language processing

Mirko Bronzi

14:45 – 15:15 : Break

15:15 – 16:30 : Tutorial

Recurrent neural networks

Access to the public github repo for the tutorials:  https://github.com/mila-iqia/ivado-mila-dl-school-2019

PROGRAM TO COME