Program

What you can expect

circle-single

Lectures and one-on-ones with world-class AI researchers

circle-single

Connect with talented researches from across 45 countries and dozens of leading industries

circle-single

Discover the latest advances in deep learning and reinforcement learning

Agenda // updates will be made as they become available

Mon Aug 3

10:00 am - 10:30 am // Welcome remarks

Pierre-Luc Bacon

Pierre-Luc Bacon

Canada CIFAR AI Chair | Mila
Université de Montréal

Yoshua Bengio

Yoshua Bengio

Canada CIFAR AI Chair
CIFAR Fellow, Learning in
Machines & Brains
Université de Montréal/Mila

Aaron Courville

Aaron Courville

Canada CIFAR AI Chair | Mila
Université de Montréal

Valerie Pisano

Valérie Pisano

President and CEO
Mila

Elissa Strome

Elissa Strome

AVP Research & Executive Director, Pan-Canadian AI Strategy
CIFAR

10:30 am - 11:45 am // Deep Commonsense Intelligence

Yejin Choi

Yejin Choi

University of Washington

11:45 am - 1:15 pm // Break

1:15 pm - 2:30 pm // Title to come

Blaise Agüera y Arcas

Distinguished Scientist
Google

2:30 pm - 2:45 pm // Break

2:45 pm - 4:00 pm // Quirks of deep emergent languages

There has recently been some interest in letting communities of neural networks evolve their own language in order to solve tasks together. The lecture will review some of this work, with an emphasis on analyzing the kind of communication code that emerges in these simulations, and what it teaches us about neural networks and languages in general.

Marco Baroni

Marco Baroni

Pompeu University
Facebook AI Research

Language Emergence

4:15 pm - 4:45 pm // Breakout Sessions with Daily Speakers

Tue Aug 4

10:30 am - 11:45 am // Intro to Deep Learning Theory

What is the effect of depth on learning dynamics in neural networks? What interplay of dynamics, architecture, and data make good generalization possible in overparameterized networks? How do deep networks organize their internal representations to represent rich structure in the world like hierarchies? This talk will give an overview of advances in deep learning theory that are beginning to shed light on these questions.

Andrew Saxe

Andrew Saxe

University of Oxford

Theory

11:45 am - 1:15 pm // Break & Speaker 1:1's

Space is very limited for speaker 1:1 sessions. Imagine a 10 minute private conversation with one of the DLRLSS speakers?

Random draws will determine the lucky few who can participate!

1:15 pm - 2:30 pm // Graph Representation Learning

In this lecture, Will Hamilton will discuss the area of graph representation learning. We will introduce standard techniques for learning low-dimensional embeddings of graph data, as well as the graph neural network (GNN) framework.

Will Hamilton

Will Hamilton

Canada CIFAR AI Chair
Mila
McGill University

2:30 pm - 2:45 pm // Break

2:45 pm - 4:00 pm // Self-supervised Visual Learning

Alexei Efros

Alexei Efros

University of California Berkeley

4:15 pm - 4:45 pm // Breakout Sessions with Daily Speakers

5:00 pm - 6:00 pm // 1:1 with speakers of the week (10-15 min, registration required. Limited number)

Wed Aug 5

10:30 am - 11:45 am // Introduction to Reinforcement Learning

This lecture provides an introduction to reinforcement learning and intelligence, which focuses on the study and design of agents that interact with a complex, uncertain world to achieve a goal. We will emphasize agents that can make near-optimal decisions in a timely manner with incomplete information and limited computational resources. The lecture will cover Markov decision processes, reinforcement learning, and function approximation (online supervised learning).

Adam White

Adam White 

Canada CIFAR AI Chair | Amii
University of Alberta, DeepMind

12:00 - 1:00 pm // Panel Discussion: What happens after the PhD? Making the jump from student to faculty member

Sarath Chandar

Sarath Chandar

Canada CIFAR AI Chair | Mila
Université de Montréal, Mila

Audrey Durand

Audrey Durand

Canada CIFAR AI Chair | Mila
Université Laval

Will Hamilton

Will Hamilton

Canada CIFAR AI Chair
Mila
McGill University

Courtney Paquette

Courtney Paquette

Canada CIFAR AI Chair | Mila
McGill University, Google

Joelle Pineau

Joelle Pineau

Canada CIFAR AI Chair | Mila
McGill University

1:15 pm - 2:30 pm // Advanced Topics in RL: Exploration & Generalization

Reinforcement learning (RL) is a systematic approach to learning and decision making. Developed and studied for decades, recent combinations of RL with modern deep learning have led to impressive demonstrations of the capabilities of today's RL systems, and have fueled an explosion of interest and research activity. This seminar starts from fundamentals of reinforcement learning and builds up to a better understanding of how domain structure and recent deep learning advances can push current limits in terms of flexible and sample-efficient reinforcement learning.

Katja Hoffmann

Katja Hoffmann

Microsoft Research

Core RL

2:30 pm - 2:45 pm // Break

2:45 pm - 4:00 pm // Intro to Bandits

This session will provide an introduction to bandits, more specifically to the stochastic bandit setting. We will review the most common algorithms, such as Upper Confidence Bound and Thompson Sampling.

Audrey Durand

Audrey Durand

Canada CIFAR AI Chair | Mila
Université Laval

4:15 pm - 4:45 pm // Breakout Sessions with Daily Speakers

5:00 pm - 6:30 pm // Trivia Night hosted by Vector

Thu Aug 6

10:30 am - 11:45 am // Model-Based Reinforcement Learning

This talk presents a broad overview of the field of model-based (deep) reinforcement learning (MBRL). MBRL methods utilize a model of the environment to make decisions and present unique opportunities and challenges beyond model-free RL. I will discuss methods for learning transition and reward models, ways in which those models can effectively be used to make better decisions, and the relationship between planning and learning. I will also highlight ways that models of the world can be leveraged beyond the typical RL setting, and what insights might be drawn from human cognition when designing future MBRL systems.

Jessica Hamrick

Jessica Hamrick

DeepMind

Model-based RL

11:45 am - 1:15 pm // Poster Presentations and Networking in Gathertown

Participants will receive an invitation link to Gathertown where posters will be virtually displayed. Move around the space and interact live with fellow participants.

1:15 pm - 2:30 pm // Reinforcement Learning via Convex Duality

We review basic concepts of convex duality and summarize how this duality may be applied to a variety of reinforcement learning (RL) settings, including policy evaluation or optimization, online or offline learning, and discounted or undiscounted rewards. The derivations yield a number of intriguing results, including entropy-regularized RL and the recently proposed *DICE family. Thus, through the lens of convex duality, we provide a unified treatment and perspective on these works, which we hope will enable researchers to better use and apply the tools of convex duality to make further progress in RL.

Offir Nachum

Offir Nachum

Google

Saddle-point optimization in RL

2:30 pm - 2:45 pm // Break

2:45 pm - 4:00 pm // Self-Play and Zero-Shot coordination in Hanabi

In recent years we have seen fast progress on a number of zero-sum benchmark problems in AI, e.g. Go, Poker and Dota. In contrast, success in the real world requires humans to collaborate and communicate with others, in settings that are, at least partially, cooperative. Recently, the card game Hanabi has been established as a new benchmark environment to fill this gap. In particular, Hanabi is interesting to humans, since it is entirely focused on theory of mind, i.e., the ability to reason over the intentions, beliefs and point of view of other agents when observing their actions. This is particularly important in applications, such as communication, assistive technologies and autonomous driving.

This talk will provide an update on recent progress in this area. It will start out with novel state-of-the-art methods for the self-play setting. Next, it will introduce the Zero-Shot Coordination setting as a new frontier for multi-agent research. Finally it will introduce Other-Play as a novel learning algorithm, which allows agents to coordinate ad-hoc and biases learning towards more human compatible policies.

Jakob Foerster

Jakob Foerster

Canada CIFAR AI Chair
Vector Institute
Facebook AI Research

Multi-agent RL

4:15 pm - 4:45 pm // Breakout Sessions with Daily Speakers

5:00 pm - 6:00 pm // Ethics in AI: Panel Discussion with Q & A

Yoshua Bengio

Yoshua Bengio

Canada CIFAR AI Chair
CIFAR Fellow, Learning in
Machines & Brains
Université de Montréal/Mila

Doina Precup

Doina Precup

Canada CIFAR AI Chair | Mila
CIFAR Fellow, Learning in Machines & Brains
McGill University

Fri Aug 7

10:30 am - 11:45 am // Safe and Data-Efficient Reinforcement Learning for Real Robots

Cras mattis consectetur purus sit amet fermentum. Cras mattis consectetur purus sit amet fermentum. Praesent commodo cursus magna, vel scelerisque nisl consectetur et. Morbi leo risus, porta ac consectetur ac, vestibulum at eros. Maecenas sed diam eget risus varius blandit sit amet non magna.

Angela Schoellig

Angela Schoellig

Canada CIFAR AI Chair | Vector Institute
University of Toronto

Robotics

11:45 am - 1:15 pm // Break

1:15 pm - 2:30 pm // Title to come

Rupam Mahmood

Rupam Mahmood

Canada CIFAR AI Chair | Amii
University of Alberta

Robotics

2:30 pm - 2:45 pm // Break

2:45 pm - 4:00 pm // Meta Reinforcement Learning

Chelsea Finn

Chelsea Finn

CIFAR Fellow, Learning in Machines & Brains
Stanford University

Transfer/Meta RL

4:15 pm - 4:45 pm // Breakout Sessions with Daily Speakers

5:00 pm - 6:00 pm // Improve Hosted by Amii

testimonial-yoshua-bengio

“The #DLRL Summer School is a place of learning at the highest scientific level in world-changing research areas but also an opportunity for networking, creating connections which can stay for life and fuel future collaborations.”

Yoshua Bengio

CIFAR Learning in Machines & Brains
Canada CIFAR AI Chair; Mila; Université de Montréal
testimonial-blake-richards

“The DLRL Summer School is an unparalleled opportunity for these young people to both learn from, and network with, the scientists who can help them launch their careers in AI.”

Blake Richards

CIFAR Learning in Machines & Brains;
Canada CIFAR AI Chair, Mila, McGill University
testimonial-hugo-larochelle

“Thanks to DLRL Summer School, I extended my personal network by meeting other student attendees like Raquel Urtasun, Mark Schmidt and Frank Wood, who are now all profs at Canadian universities and world-renowned AI researchers.”

Hugo Larochelle

CIFAR Learning in Machines & Brains;
Canada CIFAR AI Chair, Mila, Université de Montréal
testimonial-graham-taylor

“Beyond the learning and networking opportunities, [the #DLRL] showcases the rich AI ecosystem we enjoy in Canada. It's been a major factor in encouraging some of the brightest minds to launch their research careers here.”

Graham Taylor

CIFAR Learning in Machines & Brains;
Canada CIFAR AI Chair, Vector Institute, University of Guelph
testimonial-alona-fyshe

“[The #DLRL] is a wonderful opportunity to bring the AI leaders of tomorrow together for a fun and immersive learning experience. It’s also a great way to showcase the depth of AI knowledge and expertise that exists in Canada.”

Alona Fyshe

CIFAR Learning in Machines & Brains;
Canada CIFAR AI Chair, Amii, University of Alberta
circle-single-110px