Workshop on Recommender Systems

August 20-23, 2019 | MONTREAL



Recommender systems find patterns in user behaviour to improve personalized experiences and understand the environment that they are acting in. They are ubiquitous and are most often used to recommend items to users (for example, books and movies on Amazon and Netflix, relevant documentation in large software projects, or papers of interest to scientists).

The workshop on recommender systems will be held in English. This workshop includes theoretical sessions in the morning and hands-on sessions in the afternoon.

Generic objectives

  • Train industry professionals and students in the fields of data science, machine learning and operational research;
  • Develop a better understanding of the challenges and problems in data science applied to a specific field;
  • Learn to use software tools to solve practical industry-related problems;
  • Offer a networking opportunity between students and professionals from industry;
  • Facilitate knowledge transfer between academia and industry.


Basic knowledge of mathematics and programming (ideally Python) is strongly recommended.

Date and place

August 20-23, 2019
HEC Montréal – Pavillon CSC
3000, chemin de la côte Sainte-Catherine, salle Procter et Gamble
Montréal, QC H3T 2A7



  • Laurent Charlin, HEC Montréal (and Mila)
  • Brian Moore, IVADO


For additional questions or inquiries, please contact

Confirmed speakers

David Berger

David Berger is a PhD student in computer science at the Université de Montréal…

Laurent Charlin

Laurent Charlin is an assistant professor of artificial intelligence at HEC Montréal and a member of Mila–Quebec Artificial Intelligence Institute.

Didier Chételat

Didier Chételat is a researcher at the Canada Excellence Research Chair in Data Science for Real-Time Decision Making…

Fernando Diaz

Principal Researcher and lead of the Montreal FATE Research Group at Microsoft

Michael Ekstrand

Michael Ekstrand is an assistant professor in the Department of Computer Science at Boise State University.

Jonathan Guymont

Jonathan Guymont has a B.Sc degree in mathematics from Université de Montreal.

Dawen Liang

Dawen Liang is a senior research scientist at Netflix, working on core personalization algorithms

James McInerney

James McInerney is a senior research scientist at Netflix Research focusing on causality and Bayesian inference for recommendation.

Bhaskar Mitra

Bhaskar Mitra is a Principal Applied Scientist at Bing in the Microsoft Research Montreal lab



09:00-12:00: Presentation

Introduction to machine learning and deep learning

Jonathan Guymont

12:00-13:00: Lunch (included)

13:00-16:00: Hands-on tutorial

Machine learning in action (Bring your own laptop!)

David Berger (Université de Montréal) & Didier Chételat (Polytechnique Montréal)

There will be a coffee break in the morning and afternoon.


09:00-12:00: Presentation

Recommender systems basics and deep learning for recommender systems

Laurent Charlin (HEC/Mila)

12:00-13:00: Lunch (included)

13:00-14:30: Presentation

Learning to rank

Bhaskar Mitra (Microsoft Research)

15:00-17:00: Hands-on tutorial

Recommender systems in action (Bring your own laptop!)

There will be a coffee break in the morning and afternoon.


09:00-12:00: Presentation

Contextual Bandits in recommender systems

James McInerney (Netflix)

12:00-13:00: Lunch (included)

13:00-16:00: Presentation

Fairness in recommender systems

Michael Ekstrand (Boise State)

There will be a coffee break in the morning and afternoon.


09:00-12:00: Presentation

Advanced modelling

Dawen Liang (Netflix)

12:00-13:00: Lunch (included)

13:00-16:00: Presentation

Evaluating recommender systems

Fernando Diaz (MSR)

There will be a coffee break in the morning and afternoon.


Nous travaillons actuellement à distance afin de maintenir nos activités.
Formations et événements en présentiel reportés : des informations spécifiques à chaque cas seront communiquées.

We are currently working remotely in order to maintain our activities.
Face-to-face training sessions and events are postponed: specific information will be communicated in each case.