Reasoning and Learning Lab

Meeting Schedule

Time: Mondays, 11:45 am - 1:00 pm

Location: McConnell 103

Fall 2017

Date

Name

Topic

September 11

?

September 18

Emmanuel

Independently-Controlled Features

September 25

Matt

Options

October 2

Sid

Meta-Learning

October 9

No meeting

Happy Thanksgiving!

October 16

Ryan/Jean

OpenAI work (Multiagents)

October 23

Visiting: Nicholas Denis

TBD

October 30

Prasanna

Dialogue Systems

November 6

Eric

RL

November 13

Vincent

Conditional Computations & RL

November 20

Josh

November 27

Pierre

December 4

Jad

TBD

December 11

Guillaume

RL

Summer 2017

Date

Speaker

Topic

May 11

Julien Audiffren

May 18

No talk (NIPs)

May 25

 Emmanuel

June 1

Pierre-Luc

June 8

 Danlan

June 15

No Talk (RLDM)

June 22

Adam Oberman

June 29

No Meeting

July 6

Jackie & Pascale

July 13

Di Wu

July 20

Hossein

July 27

Visiting: Christopher Maddison

August 3

Herke

August 10

Sanjay

August 17

No talk

August 24

Undergraduate Presentations

August 31

 Herke

Winter 2017

Date

Speaker

Topic

Jan 12

 NIPS review

Jan 19

Pierre-Luc

Jan 26

 Guest (Ross Otto?)

Feb 2

 Matt

Feb 9

 Emmanuel & Vincent

Feb 16

 Leo

Feb 23

 Herke

Mar 2

 Reading week (no meeting)

Mar 9

 Maziar

Mar 16

 Sean Smithson

Mar 23

 Neil

Mar 30

 Danny Tarlow

Apr 6

 Martin

Apr 13

 Charles & Pierre-Luc

Apr 20

 Pierre & Pascal + Matt

Apr 27

 Guillaume

May 4

 Sumana

Fall 2016

Date

Speaker

Topic

September 8

Organizational meeting

September 15

Sydney Swaine-Simon

Harm van Seijan (guest: Maluuba)

The Watson AI XPrize

True Online TD

September 20 (special invited talk)

Pascal Poupart (visitor from U of Waterloo)

Leveraging Data Science to Directly Learn Tractable Models for Probabilistic Inference and Decision Making

September 22

Guillaume Rabusseau (visitor)

Tensors, Weighted Automata and Reduced-Rank Regression

September 29

Emmanuel, Ryan and Josh

Recent papers of interest

October 6

cancelled

October 13

Charles

Prediction of extubation readiness in preterm newborns based on analysis of cardio-respiratory metrics

October 19: special meeting at 10 AM

Boyu

Transfer and Multitask Learning

October 20

Eric

October 27

Pascale

November 3

Maziar, Ryan and Emmanuel

recent papers

November 10

Dialogue Team (Ryan, Michael, ...)

November 17

Maziar

November 24

Doina

December 1

Di

December 8

Cancelled: NIPS

December 15

2:00 in  Arts 215

Vince Conitzer

Summer 2016

Date

Speaker

Topic

May 12

none

Organizational Meeting

May 19

ICLR recap

May 26

Pierre

Seizure detection

June 2

Pascale

WFA

June 9

Pierre-Luc

Options

June 16

Maziar

Differential privacy in rl

June 23

Jad

DL for definiteness prediction

June 30

ICML recap

July 7

Josh & Emmanuel

Conditional computation

July 14

cancelled

July 21

Ryan

Deep learning for dialogue systems

July 28

Harsh

Real-time machine translation

August 4

cancelled

August 8

Visitors:

Anna Harutyunyan

Craig Sherstan

Off-policy TD Learning from Returns

Prosthetics, Predictions and Confidence

August 9

Visitors:

Ozlem Aslan

Nadia Ady

Global, Robust and Compact Representation Learning

Computational Curiosity

August 18

cancelled

August 25

Kian

Michael

Philip

Johanes

Winter 2016

Date

Speaker

Topic

January 20

Jad

Deep learning and graphical models

January 27

Emmanuel Bengio

Conditional computing

February 3

Jean Harb

Intrinsic motivation/variational inference

February 10

Pierre-Luc

Option-critic, eligibility traces, bounded rationality

February 17

guest

February 24

Priya

NLP

March 9

March 16

Boyu

Multi-task/transfer learning

March 23

Eric

March 30

Jean

Value Iteration Networks

April 6

Phil

Controlling complexity by sharing parameters and minimizing variation

April 13

Alex

Nested Dirichlet Process Applied to Censored Data

April 20

April 27

Herke van Hoof (guest)

Robot Learning and Exploration in Sensory Rich Environments

Fall 2015

Date

Speaker

Topic

September 9th

Organizational meeting

September 16th

Pierre-Luc

The options-critic architecture

September 23rd

Maziar

September 30th

Melanie

October 7th

Phil

October 14th

Emmanuel

Theano tutorial

October 21st

Leo

October 28th

Martin

November 4th

Laurent

November 11th

Neil

November 18th

Jad

November 25th

Eric

December 2nd

Visitors

December 14th

Philip Thomas

Reinforcement learning

December 16th

NIPS Recap / Borja

Summer 2015

Date

Speaker

Topic

May 5th

Francois Rivest

Learning timing should be easy

May 11th

Suchi Saria

Learning Models for Monitoring and Prognonis for Electronic Health Data

May 25th

Tony Jebara

Graphical modelling with the Bethe approximation

May 27th

Prakash

Bisimulation

May 29th

Lab BBQ

June 3rd

Joelle

Improving the design and discovery of dynamic treatment strategies using recent results in sequential decision-making (ICAPS 2015)

June 10th

Borja

A canonical form for Weighted Automata

June 13th

Rival lab BBQ

June 17th

Pierre-Luc

Importance sampling for off-policy learning  (RL)

June 23rd

Ryan

ICLR/RLDM recap

June 30th

Phil

Data imputation as guided policy search (RL+DL)

July 15th

Gheorghe

Basis refinement for linear value function approximation (RL)

July 22nd

Eric

Spectral, IRL

July 29nd

ICML discussion

August 5th

Martin

Learning from demonstration for the SmartWheeler

August 12th

Melanie

Matrix completion in PSR learning

August 19th

UGs

August 26th

UGs

Winter 2015

Date

 

Speaker

Topic

 

Abstract/Resources

28.01

NIPS Review

04.02

Cosmin Paduraru

MDPs for mineral extraction and processing

https://sites.google.com/site/cosminpaduraru/publications

11.02

Audrey Durand

An overview of bandit problems

http://vision.gel.ulaval.ca/en/people/Id_640/index.php

18.02

Laurent Charlin

Recommender systems

http://www.cs.toronto.edu/~lcharlin/

25.02

Philip Bachman

 Training generative models

ICML 2015

Reading week

11.03

Jackie Chi Kit Cheung

Distributional semantics and natural language generation

http://cs.mcgill.ca/~jcheung/

18.03

Cancelled

25.04

Ryan and Nissan

Dialogue datasets and neural language models

01.04

Cancelled

08.04

Cancelled

15.04

Pierre-Luc Bacon

Learning recognizers

Barbados 2015

22.04

TBD

Fall 2014

Date

 

Speaker

Topic

 

Abstract/Resources

09.09

Faizy Ahsan

Density-Based Clustering and Intrinsic Dimensionality

09.09

Angus Leigh

Laser-based person tracking for clinical locomotion analysis

16.09

Eric Crawford

Biologically Plausible,        Human-scale Knowledge Representation

23.09

Emmanuel Bengio

Deep learning

30.09

Rich Sutton

Learning to predict independent of span

07.10

Pierre-Luc

Spectral Learning and Options

21.10

Boyu Wang

Online learning algorithms for transfer and multitask learning

11.11

Andrew + Angus

18.11

Borja + Pierre-Luc

25.11

Nastaran

Machine Learning for Disease Outbreak Detection

02.12

Martin (tentative)

09.12

NIPS

16.12

Past - Summer 2014

Date

 

Speaker

Topic

 

Abstract/Resources

27.08

Gab

20.08

UGRAD

13.08

UGRAD

06.08

AAAI Recap

30.07

23.07

Mahdi

16.07

Jin Xu

09.07

ICML Recap

02.07

P-L

25.06

Clement

18.06

Gheorghe

11.06

Phil

ICML practice talk

04.06

Kim

Activity recognition on the wheelchair using accelerometer data

28.05

Phil

Pseudo-ensembles and dropout

21.05

Andrew

Recap on field trials

14.05

Robert

Application of reinforcement learning for medical problems

Past - Spring 2014

Date

 

Speaker

Topic

 

Abstract/Resources

12.02

Mahdi Milani Fard

Cancelled

19.02

William Hamilton

A tutorial on tensor decomposition for learning latent variable models

26.02

Yuri Grinberg

Optimization of hydro-electric power plant using (RL and PSRs)

05.03

Study break

12.03

Jinxu Jia

Cancelled

19.03

Borja Balle

Efficient Learning Algorithms for Weighted Automata (Beyond Spectral Methods for Generative Models)

26.03

Prof. Doina Precup

Cancelled

02.04

Yuri Grinberg

Modeling With Conditional Predictive State Representations

09.04

Ouais Alsharif

Lifelong Learning of Discriminative Representations

16.04

Martin Gerdzhev

Cancelled

23.04

Pierre-Luc Bacon

Subjective localization with options maps

Slides here

30.04

Gheorghe Comanici

For information, contact: gcoman@cs.mcgill.ca

Past presentations

Date

 

Speaker

Topic

 

Abstract/Resources

05.02

The Smart Wheelchair Group

22.01

Phil Bachman

Clement Gehring

15.01

Prof. Prakash Panangaden

2013

12.17

Negar Ghourchian

Mobile Data Analysis Using Hierarchical Dirichlet Process

12.10

Hang Ma

Information Gathering and Reward Exploitation of Subgoals for POMDPs

12.03

Borja Balle

Spectral methods

11.19

Timothy Mann

The Advantage of Planning with Options

11.14

Gabor Lugosi

Prediction and online combinatorial optimization

10.29

Borja Balle

An Introduction to Random DFA

11.05

Nastaran Jafarpour

Using Hierarchical Mixture of Experts model for fusion of outbreak

detection methods

11.12

Ladan Mahabadi

Music Self-similarity and complexity leveraged for composer classification and computational Turing tests

10.22

Amir massoud Farahmand

Beyond the regularities of the Value Function

10.08

Pierre-Luc Bacon

Mixture of options

10.01

Martin Gerdzhev

An enhanced system for augmenting urban search and rescue canines

09.24

Ouais Alsharif

End-to-end Text Recognition using Hybrid HMM Max-out Models

09.17

Phil Bachman

Greedy Confidence Pursuit

Abstract

09.10

Hamid Reza Maei

Learning about a good policy from data generated according to some decision policies

Abstract

08.28

Bénédicte Leonard-Cannon

Lucas Lehnert

Adam Bene Watts

Undergraduate Research Presentations

Abstracts

08.21

Donald Macisaac

Undergraduate Research Presentations

07.31

Prof. Joelle Pineau

Tutorial on POMDPs

07.24

Yuri Grinberg,

Prof. Joelle Pineau

ICML Recap

Conf. Website

Videos

07.10

Ouais Alsharif

Deep belief nets

Paper

06.19

Artem Kaznatcheev

Weighted automata are compact and actively learnable

05.29

Yogesh Girdhar

Topic Modeling for Robots

Abstract

05.22

Will Hamilton

Modelling Sparse Dynamical Systems with Compressed PSRs

Abstract

05.22

Raheem Adam

Arcade learning environments

05.15

Pierre-Luc Bacon, Clement Gehring

AAMAS recap.

AAMAS2013 official site

Slides

Pierre-Luc Presentation

04.24

Andre Barreto

Practical guide to Kernel-Based Stochastic Factorization

Abstract

04.03

Pierre-Luc Bacon

Using Label Propagation for Learning Temporally Abstract Actions in Reinforcement Learning

Abstract

03.27

Clement Gehring

Smart Exploration in Reinforcement Learning using Absolute Temporal Difference Errors

Abstract

03.13

Negar Ghourchian

Place Identification and Prediction in the D4D Data Set using Machine Learning

Abstract

02.27

Beomjoon Kim

Imitation Learning for Robot Navigation

Abstract

02.20

Boyu Wang

Online Ensemble Learning for Imbalanced Data

Abstract

02.13

Guillaume Saulnier Comte

Machine Learning Toolbox for Automating the Development of Personalized Epileptic Seizure Detection Algorithms.

Abstract

02.06

Shaun Zia

Cybernetics and the Autistic Mind: A Reductionist Approach to Understanding Perception and Cognition

Abstract

01.30

Amir-massoud Farahmand

Reinforcement Learning Problems, their Regularities, and Nonparametric Algorithms to Solve them

Description

01.23

Philip Bachman

Greedy confidence pursuit: A Pragmatic Approach to Multi-Bandit Optimization

Abstract

01.16

Sara Marie McCarthy

A Theoretical Analysis of Variable Actions for Random Walks in N Dimension

Abstract

01.16

Gheorghe Comanici

Personalizing Education using Machine Learning

NIPS Workshop

Paper

coursera.org

2012

12.19

Athena

12.12

Andre + Arthur

NIPS recap

11.28

Nastaran

11.21

Andre

11.14

Cancelled

11.07

Mohammad (INRIA)

10.31

Sylvie

10.24

Cancelled

10.17

Mahdi

10.10

Amir massoud

10.03

Cosmin

09.26

Neil

09.19

Marc

09.12

Gheorghe

08.28

Undergraduate student presentations

08.21

Undergraduate student presentations

08.14

Danesh

08.7

Prakash

07.31

Atoussa

07.24

Cosmin

07.17

Phil/Mahdi

ICML recap + AAAI practice talk

07.10

Andre

moved to TTT

07.3

cancelled

06.26

Nastaran

06.19

Phil

ICML paper presentation

06.12

Negar

06.5

Yuri

AISTATS recap

05.29

Prakash

TBD

05.22

Prakash

TBD

05.15

Prakash

TBD

05.7

Pierre-Luc (short talk) + planning the summer

05.1

Hanna Kurniawati

A POMDP approach for global motion planning...

04.24

Jordan

04.17

cancelled

04.10

Phil

reading group on Online Learning & Bandits

04.3

Mahdi

reading group on Online Learning & Bandits

03.27

Konstantinos Tsianos

reading group on Online Learning & Bandits

03.20

Amir Massoud

reading group on Online Learning & Bandits

03.13

Bert

03.6

Cosmin

02.28

cancelled

02.21

study break

02.14

Phil

02.7

Arthur

01.31

Yuri & Andre

NIPS recap

01.24

cancelled

01.17

Joelle & Amir Massoud

NIPS recap

2011

09.12

 

Mahdi

 

08.24

 

Undergrad presentations

 

08.17

 

Undergrad presentations

 

08.10

 

Amir massoud

 

08.3

 

Gheorghe

AAAI Practice Talk

 

07.27

 

Mitchel (Neural Prosthetics Lab)

 

07.20

 

Cancelled

Cancelled

 

07.13

 

Andre

 

04.27

 

Athena

 

04.20

 

Danesh / Nakisa

 

04.13

 

Sylvie

 

03.30

 

Bert

 

03.16

 

Guillaume / Jordan

 

03.9

 

Gheorge

 

03.2

 

Cosmin

 

02.16

 

Shaowei

IPPC

 

01.19

 

Sylvie

POMDP

 

01.12

 

Mahdi / Jordan

NIPS Summary

 

2010

12.1

 

Mahdi

 

11.24

 

Phil

 

11.17

 

Sylvie

 

11.10

 

Shaowei

 

11.3

 

Cosmin

 

10.27

 

Bert

Model-free Bayesian RL

 

10.20

 

Jordan

UbiComp Summary

 

10.13

 

Gheorghe

Spectral clustering for FA

 

10.6

 

Susan

Semi-parametric

 

09.29

 

Yuri

Sequential methods

 

09.22

 

Doina

LSTD / AAAI summary

 

09.15

 

Pablo

Julieta

ECML practice talk

Machine-Learning Techniques to Optimize Neurostimulation Strategy in Epilepsy Treatments

 

09.8

 

Andre

Stochastic factorization

 

08.25

 

Guillaume Saulnier/Jonathan Cottrell

Summer Student Presentations

 

07.07

 

Yuri Grinberg/Doina Precup

ICML Recap/AAAI Practice #2 Talk

 

06.30

 

Doina Precup/Pablo Castro

AAAI Practice Talks

 

06.23

 

Cancelled

Cancelled

 

06.16

 

Yuri Grinberg

ICML Practice Talk

 

06.09

 

Mahdi Milani Fard

PAC Bayes RL

 

06.02

 

Amin Atrash

Wheelchair Project

 

05.26

 

Pablo Castro, Rob/Arthur

Interview Practice

 

05.19

 

Pablo/Gheorgh

AAMAS Debrief

 

05.12

 

Robert Vincent

TBD

 

05.05

 

Amir-massoud Farahamand, U of A

Guest Lecture

 

04.28

 

Arthur Guez/Robert Kaplow

TBD

 

04.21

 

Cancelled

TBD

 

04.13

 

Susan Shortreed

TBD

 

04.06

 

NO SEMINAR

 

03.30

 

Ryan Faulkner

Deep-belief Networks

 

03.23

 

Yuri Grinberg

Transformed PSRs

 

03.16

 

Doina Precup

TBD

 

03.09

 

Shao-wei Png

Activity Recognition

 

03.02

 

Jordan Frank

Activity Recognition

 

02.16

 

Doina Precup/Keith Bush

Talk and Planning Session

 

02.02

 

Cancelled

 

02.02

 

Gheorge Comanici

Options

 

01.26

 

Mahdi Milani Fard

PAC-Bayes

 

01.19

 

Keith Bush

Context-based Interaction

 

01.12

 

Phil Bachman

TBD

 

01.05

 

Marc Bellemare

TBD

 

2009

12.15

 

Keith Bush, Arthur Guez, Mahdi Milani Fard, and

Gheorghe Comanici

NIPS after-action debrief

 

12.08

 

Amin Atrash

Wheelchair Project

 

12.01

 

Ivan Savov

Intro to Quantum Computing

 

11.24

 

Monica Dinculescu

Mystery Title

 

11.17

 

Francois Rivest

(RL Mechanisms in real animals)

 

11.10

 

Keith Bush

Planning Session

 

11.03

 

Cancelled

MC103 used for exam

 

10.29

 

Doina Precup

Fast Gradient Methods for TD

 

10.20

 

Robert West

TBD

 

10.13

 

Doina Precup

TBD

 

10.06

 

Cosmin Paduraru

Cross-validation of Batch RL Methods

 

09.29

 

Keith Bush

Model-based Reinforcement Learning for Neurostimulation

 

09.22

 

Jordan Frank

Compressed Sensing

 

09.01

 

Arthur Guez

Simultaneous Localization and Mapping (SLAM)

 

08.25

 

Olivier Remillard, Guillaume Saulnier, and Marcos Ginestra

Undergraduate Summer Research Presentations

 

08.18

 

Robert Vincent

Continuous Action Spaces (Discussion)

 

Paper 1

Paper 2

08.11

 

Vacation

 

08.04

 

Vacation

 

08.28

 

Vacation

 

07.21

 

Vacation

 

07.14

 

Mahdi Milani-Fard

Non-deterministic Policies in Markovian Processes (Make-up)

 

07.07

 

Andre' da Motta Salles Barreto

(guest lecture)

Approximate Dynamic Programming Using the Stochastic Factorization

 

Abstract

06.30

 

Robert West/Pablo Castro

Wikispeedia/POMDPs

 

06.23

 

Marc Deisenroth (guest lecture)

Probabilistic Inference for Fast Learning in Control

Abstract

06.16

 

ICML/UAI/COLT

 

06.09

 

ICML/UAI/COLT

 

06.02

 

Mahdi Milani Fard

Postponed

 

05.26

 

Pablo Castro

Practice Talk

 

05.19

 

Julien Villemure

SLAM

 

05.12

 

Robert Vincent

Computational Modeling of Epilepsy

 

05.05

 

CANCELLED

 

04.28

 

Doina Precup

Gradient-descent TD methods for function approximation

 

04.21

 

Sheryl Morrissey

Minerva Expense Reporting Tutorial

 

04.14

 

CANCELLED

 

04.07

 

Monica Dinculescu

Learning approximate predictive models

 

Abstract

03.31

 

Phillipe Chapu

Approximating Markov processes by averaging

 

Abstract

03.24

 

Keith Bush

Manifold Embeddings for Reinforcement Learning with Partial Observability

 

Abstract

03.17

 

Cosmin Paduraru

Model-based Bayesian Reinforcement Learning with Adaptive Discretization

 

Abstract