Schedule of the SequeL Weekly Group Meeting - Winter & Spring 2013

All meetings will be held in room W01 at 11h00.

Jan. 11
Pierre Chainais On Bayesian Non-Parametric Methods and Image Processing (part 1)
Jan. 18
Pierre ChainaisOn Bayesian Non-Parametric Methods and Image Processing (part 2)
Jan. 25
Emmanuel Duflos Dirichlet process mixture for joint state-noise pdf Bayesian estimation in dynamical systems
Feb. 1
Balazs Szorenyi
Adapting multi-armed bandits to peer-to-peer networks
Feb. 6
Joelle Pineau

Feb. 8
Pierre Chainais On Bayesian Non-Parametric Methods and Image Processing (part 3)
Feb. 22
Romaric Gaudel
Efficient Eigen-updating for Spectral Graph Clustering
Feb. 28
Masrour Zoghi
Relative Thompson Sampling
Mar. 1
Gunnar Kedenburg
Learning and Planning tasks for Wind Turbine Control
Mar. 5
Jennifer Healey
Sensing Emotion in the Wild
Mar. 8
Hai Nguyen Reliable machine learning algs. and application to information security and digital forensics
Mar. 22
Emilie Kaufmann
Apr. 5
Adam Bull
Adaptive-treed Bandits
May 17
Giovanni Zappella
Sequential Learning Problems on Graphs
June 14
Ludovic Denoyer
Learning with relational data: sequential models and propagation models for
structured classification and labeling

Schedule of the SequeL Weekly Group Meeting - Fall 2012

All meetings will be held in room W01 at 11h00.

Sep. 14
Nicolas Drougard
Possibilistic MDP and POMDP
Sep. 21
Remi Coulom
Automatic Go Board Image Recognition
Sep. 26
Joelle Pineau Bayesian Planning and Learning with Uncertain Models
Sep. 28
Joelle Pineau
Automatic Detection and Suppression of Epileptiform Activity using
Reinforcement Learning
Oct. 12
Ronald Ortner
Regret Bounds for Colored and Continuous State MDPs
Oct. 19
Madalina Drugan Multi-objective reinforcement learning: new techniques and analysis
Oct. 22
Gusztav Morvai
Universal Tests for Memory Words
Oct. 25
Toby Dylan Hocking
Learning sparse penalties for change-point detection using max-margin interval regression
Oct. 26
Daniil Ryabko
Unsupervised Learning of Representations that Preserve Time-series Dependence
Nov. 2
Tor Lattimore
PAC Bounds for General Reinforcement Learning
Nov. 9
Azadeh Khaleghi
Locating Changes in Highly-Dependent Data with Unknown Number of Change Points
Nov. 23
Bruno Scherrer
On the use of non-stationary policies for stationary infinite-horizon Markov decision processes
Nov. 26
Vladimir Novak
Restless bandit index policy for scheduling of queues with abandonments
Dec. 18
Ali Ghodsi
Non-Linear Dimensionality Reduction by Isometric Patch Alignment
Dec. 21
L. A. Prashanth
Stochastic Optimization Methods in Vehicular Traffic Control and Service Systems

Schedule of the SequeL Weekly Group Meeting - Winter & Spring 2012

All meetings will be held in room W01 at 11h00.

Jan. 25
Masashi Sugiyama Density Ratio Estimation: A New Versatile Tool for Machine Learning
Jan. 27
Nathaniel Korda
Finite Time Analysis of the Thompson Algorithm (
Jan. 30
Yassin Abbasi
Online-to-Condence-Set Conversions and Application to Sparse Stochastic Bandits
Feb. 3
Ronald Ortner Online Aggregation in Reinforcement Learning
Feb. 10
Remi Coulom
CLOP: Confident Local Optimization for Noisy Black-Box Parameter Tuning
Mar. 2
Phuong Nguyen
Context Tree-Maximizing Reinforcement Learning
Mar. 23
Amin Mantrach
A General Framework for People Retrieval in Social Media with Multiple Roles
Mar. 30
Azadeh Khaleghi
Online Clustering of Processes
Apr. 6
Olga Klopp
Completion of Small Rank Matrices from Noisy Observations
Apr. 27
Wouter Koolen
Buy low, sell high
June 8
Raphael Fonteneau
Min-max Generalization for Deterministic Batch Mode Reinforcement Learning
June 13 - 14h00
Céline Duval Statistical Inference Across Time Scales
June 18 - 14h00
Michael Johanson
Achieving Professional-Level Strength in Computer Poker
June 22
Branislav Kveton
Kernel-based Reinforcement Learning on Representative States

Schedule of the SequeL Weekly Group Meeting - Fall 2011

All meetings will be held in room W01 at 11h00.

Sep. 23
Odalric Maillard
Apprentissage séquentiel: Bandit, Statistique et Renforcement
Sep. 30
Michal Valko Online Semi-Supervised Learning on Quantized Graphs
Oct. 7
Matthieu Geist
Sparsity through Regularization for TD-Learning
Oct. 14
Lucian Busoniu
Optimistic Planning in Markov Decision Processes
Oct. 21
Victor Gabillon
Multi-Bandit Best Arm Identification
Oct. 28
Alexandra Carpentier
Finite Time Analysis of Stratified Sampling for Monte Carlo
Nov. 4
Alessandro Lazaric
Introduction to Mechanism Design and its Connection to Machine Learning
Nov. 18
Rémi Munos
Optimistic Optimization of a Deterministic Function without the Knowledge of
its Smoothness
Nov. 25
Damien Ernst
Learning for Exploration-Exploitation in RL: The Dusk of the Small Formulas' Reign
Dec. 2
Abdeslam Boularias
Learning Robot Grasping with Structured Output Prediction Techniques
Dec. 9
Michal Valko
Submodularity and Submodular Minimization in the Bandit Setting

Schedule of the SequeL Weekly Group Meeting - Winter & Spring 2011

All meetings will be held in room W01 at 11h00.

Jan. 7
Robby Goetschalckx
Modelling and planning a person during stroke rehabilitation with the use of PODMPs
Jan. 14
Nataliya Sokolovska
(joint meeting with Mostrare)
Jan. 21
Christophe Salperwyck
Effect of the training set size: an empirical study on leading classifiers
Feb. 4
Lei Yu
Compressive Sensing and Sparse Bayesian Learning with Cluster Sparsity Patterns
Feb. 11
Romaric Gaudel
Paramètres d'ordre et Sélection de Modèle en Apprentissage Automatique
Feb. 18
Alexandra Carpentier
Upper-Confidence-Bound Algorithms for Active Learning in Multi-Armed Bandits
Feb. 25
Odalric Maillard
Finite-time Analysis of Multi-Armed Bandit Problems with Kullback-Leibler Divergence
Mar. 4
Brahim Chaib-draa
Game Theory as a Formal Model for Learning and Planning in Multiagent Environments
Mar. 11
Victor Gabillon
Classification-based Policy Iteration with a Critic
Mar. 18
Olivier Pietquin
Kalman Temporal Difference Learning
Mar. 25
Hachem Kadri
Operator-valued Kernels for Nonparametric Operator Estimation
Apr. 1
Simon Bernard
Random Forests : Dynamic Induction and One-Class Learning (joint with Mostrare)
Apr. 8
Julien Perez
Sequential Learning for Large-Scale Streaming Dataset, a Case Study of Affinity
Propagation Clustering and Gaussian Processes Regression (joint with Mostrare)
Apr. 11
Yuval Tassa
Recasting optimal control problems with periodic solutions as inference problems
over a ring-shaped graphical model
Apr. 15
Laurent Hartet
Learning and Pattern Recognition in Dynamic Environments Applied to the
Monitoring of Evolving systems (joint with Mostrare)
Apr. 22
Stéphane Gaiffas
High-Dimension Matrix Prediction
May 4
Antoine Salomon
Robustness of Anytime Bandit Policies
May 6
Joris Bierkens
Learning and Controlling a Markov Process through its Symbol
May 27
Emilie Kaufmann
Stochastic Bandits: A Bayesian Approach
June 10
Nathan Korda
Optimistic and Ergodic Bandits
June 24
David Silver

Schedule of the SequeL Weekly Group Meeting - Fall 2010

All meetings will be held in room W01 at 11h00.

September 10
Alexandra Carpentier
Brownian sensing
September 17Peter Auer
Near-optimal regret bounds for reinforcement learning
September 24
Pierre Cheinais
Self-similar stochastic processes to model disordered systems
October 1
Lucian Busoniu
Approximate dynamic programming and reinforcement learning for
October 8
Matt Hoffman
Planning as inference
October 15
Rémi Munos
Least-squares linear regression based on random vectorial subspaces
October 22
Mohammad Gheshlaghi Azar
Dynamic policy programming: A direct policy-search approach to
Markov decision problems
November 5
Rémi Munos
Finite-sample Analysis of Bellman Residual Minimization
November 19
François Caron
Efficient Bayesian Inference for generalized Bradley-Terry Models
November 26

December 3

December 10
NIPS - 2010
December 17
Sébastien Gerchinovitz
Sparsity regret bounds for individual sequences in online linear regression

Schedule of the SequeL Weekly Group Meeting - Winter & Spring 2010

All meetings will be held in room W01, except the very first one on January 15, which will be in W21.

January 15Boris Iolis
Game Tree Search Strategies for Computer Poker
January 22Thomas Laloe
Statistical learning: classification, regression, and applications
January 29Emmanuel Duflos
Belief Functions
February 5Boris Ryabko
Asymptotically Optimal Perfect Steganographic Systems
February 12Francis Bach
High-Dimensional Non-Linear Variable Selection through Hierarchical
Kernel Learning
February 19Rémi CoulomParameter Optimization from Binary Response with Local Cubic Regression
February 26Sébastien BubeckMinimax policies for combinatorial prediction games
March 5Alessandro LazaricAnalysis of a Classification-based Policy Iteration Algorithm (PDF)
March 12Mohammad GhavamzadehBayesian Multi-Task Reinforcement Learning (PDF)
March 19Laszlo GyorfiEmpirical Portfolio Selection
April 16Rémi MunosFinite sample analysis of LSTD
April 23Hachem KadriKernel-based Learning from Functional Response Data
April 30Daniil Ryabko
Clustering Processes (PDF)
May 7
Lukasz Lew
Simulation-based Search of Combinatorial Games
May 21
François Vialatte
EEG neurodynamic analysis, models and scientific aspects
June 11
Philippe Berthet
Strong Invariance Principle and Coupling
June 22
Vincent Lemaire
A method to build a representation using a classifier and its use in a K Nearest Neighbors-based deployment

Schedule of the SequeL Weekly Group Meeting - Fall 2009

All meetings will take place in room W11, except the very first one on September 11, which will be held in W21.

September 11Victor Gabillon
Mixing multi-armed bandits and linear programming for Internet
banner advertisement
September 18Mohammad GhavamzadehKernelized value function approximation for reinforcement learning [pdf]
September 25Odalric MaillardComplexity versus agreement for many views [pdf]
October 2Alessandro LazaricA Stochastic View of Optimal Regret through Minimax Duality [pdf]
October 9Daniil RyabkoSequence prediction inside out
October 16Rémi MunosCompressed Least Squares Regression [pdf]
October 23Sébastien BubeckOptimal rates of convergence for the simple regret
October 30Sertan GirginAn MDP-based recommender system [pdf]
November 6Manuel Loth & Philippe PreuxECON: Equi-COrrelation Networks, or LARS on large feature spaces &
LARS²: fit the noise as well as the model, to help model selection
November 13Hachem Kadri
Functional regression: an RKHS approach [pdf]
November 20Rémi GribonvalSparse data modeling and dictionary learning
November 27Laurent JacquesA Compressed Introduction to Compressed Sensing: Combining
Sparsity and Sampling [slides]
December 4Jean-François HrenOptimistic planning with continuous actions papers for reference [pdf] [pdf]