Seminar

SequeL seminars are held in room A00 on Friday mornings, at 11am (any change of schedule is reported in the Comment section). Seminars either present original work of the speaker, or part of the reading group talk series. You can also find all events in the public Google Calendar feed of the seminar series, or on the new seminars platform of the University of Lille 1. For further information, contact Emilie.

Date Speaker Title Comment
Sept 20th Bruno Scherrer (Inria Nancy) On the complexity of Policy Iteration (for Deterministic Problems) Tuesday, 2PM
Sept 30th Jean Lafond (Telecom ParisTech) On the Online Frank-Wolfe algorithms for General Low Rank Matrix Completion
Oct 14th Emile Contal (ENS Cachan) The geometry of Gaussian processes and Bayesian Optimization
Nov 4th Ronan Fruit (SequeL) Exploration-Exploitation in MDPs with Options
Nov 18th Claire Vernade (Telecom ParisTech) Insights on Multiple-Plays Bandits with Censored Feedbacks room A11
Nov 25th Ludovic Denoyer (LIP 6, UPMC) Learning Sequential Predictive Models under Budget Constraints
Nov 28th Andras Gyorgy (Imperical College) Following the Leader and Fast Rates in Linear Prediction Monday, 11.15am, room A00
Dec 16th Joon Kwon (Ecole Polytechnique) Mirror descent strategies for regret minimization and approachability
Jan 20th Daniele Calandriello (SequeL) Pack only the essentials: distributed sequential sampling for adaptive kernel dictionary learning
Feb 3rd Jaouad Mourtada Streaming aggregation of a growing number of experts
Feb 10th Pierre Gaillard (Inria Paris) Sparse Accelerated Exponential Weights
March 3rd James Ridgway (SequeL) Sampling normalizing constants in high dimension using inhomogeneous diffusions
March 10th Anna Harutyunyan (VUB) Potential-based reward shaping as a tool to safely incorporate auxiliary information
March 17th Aditya Gopalan (Indian Institute of Science) Collaborative bandits on a network
March 24th Benjamin Guedj (MODAL) A quasi-Bayesian perspective to NMF: theory and applications
March 31st Tristan Cazenave (LAMSADE) Improvements to Monte-Carlo Tree Search
April 14th Ralph Bourdoukan (SequeL) TBA

Seminars of related interest and their usual schedule:

Here is the page of our reading group.

Past seminars:

Date Speaker Title Comment
Oct 30th Akram Erraqabi Pliable rejection sampling and error-regret tradeoff for a MAB problem
Nov 5th Audrey Durand Bandits for healthcare Thursday, 2:00 PM
Nov 6th Romain Warlop Hierarchical Exploration for Accelerating Contextual Bandits reading group
Nov 12th Maluuba Presentation of the company Thursday, 10:30 AM, room A11
Nov 20th Florian Strub Deep Learning for Dummies
Nov 27th Christos Dimitrakakis Differential Privacy, Bayesian Inference and More
Jan 15th MAGNET: Mark Herbster Predicting a Switching Sequence of Graph Labelings 2pm, B21
Jan 29th Bilal Piot Score-based Inverse Reinforcement Learning
Feb 12th The NIPSers NIPS Debrief (3 papers) 10h30-12h
Feb 19th Richard Combes Learning to Rank : Regret Lower Bounds and Efficient Algorithms
Mar 3rd Christopher Dance Gittins Index Theorem and Calculating Gittins Indices Thursday, 3.30 pm
Mar 4th Christopher Dance When are Kalman-Filter Restless Bandits Indexable? Linking Whittle indices, maps-with-gaps and mechanical words.
Mar 18th Emilie Kaufmann Optimal Best Arm Identification with Fixed Confidence
Apr 1st Vianney Perchet Online Learning in repeated auctions
Apr 15th Wouter M. Koolen MetaGrad: Faster Convergence Without Curvature in Online Convex Optimization
Apr 22nd Tomáš Kocák Online learning with noisy side observations 2 p.m.
May 13th Rémi Bardenet Monte Carlo with Determinental Point Processes
May 27th Cricia Zilda Felicio Paixao Extending model-based recommenders to deal with user cold-start problem

2015 spring/summer:

Date Speaker Title Comment
Jan 8th Thibaut Munzer Inverse Reinforcement Learning in Relational Domain
Jan 16th Gergely Neu Online learning in Markov decision processes 2:00 PM
Jan 30th Hadrian Glaude Subspace Identification for Predictive State Representation by Nuclear Norm Minimization
Feb 20th Romaric Gaudel Online Matrix Completion Through Nuclear Norm Regularisation
Feb 27th Jean-Bastien Grill Learning to Optimize via Information-Directed Sampling reading group
Mar 6th Peter Grünwald Learning the learning rate: how to repair Bayes when the model is wrong salle plenière
Mar 20th Benjamin Guedj Aggregation of estimators: Theory and methods
Mar 27th Julien Perolat Conditional Swap Regret and Conditional Correlated Equilibrium reading group
Apr 2nd Pratik Gajane REX3: An algorithm for the Adversarial Dueling Bandit problem
Apr 24th Gergely Neu Exploiting easy data in online optimization
Jun 19th Marta Soare Sequential Transfer of Samples in Linear Bandit
Jun 26th Julien Perolat Approximate Dynamic Programming for Two-Player Zero-Sum Markov Games
Jun 30th Jacob Abernethy Minimax Solutions, Random Playouts, and Perturbations

2014 autumn/winter:

Date Speaker Title Comment
Oct 3rd Frédéric Guillou Ensemble Contextual Bandits for Personalized Recommendation reading group
Oct 10th Tomáš Kocák Eluder Dimension and the Sample Complexity of Optimistic Exploration reading group
Oct 24th Daniele Calandriello Sparse Multi-task Reinforcement Learning
Oct 31st Julien Perolat Strategy Iteration Is Strongly Polynomial for 2-Player Turn-Based Stochastic Games with a Constant Discount Factor reading group
Nov 4th Aivar Sootia Shaping Pulses to Control Multi-Stable Biological Systems
Nov 7th Bilal Piot Apprentissage par imitation et tranfert de tâche pour une interaction homme-machine narurelle
Nov 13th Pratik Gajane Studies in multiarmed bandits and PhD topic overview
Nov 19th Michael Bowling Algorithms for computing game theoretic solutions to extremely large extensive games
Nov 21st Timothé Collet Active Learning for Classification: An Optimitic Approach
Nov 28th Tomáš Kocák Efficient learning by implicit exploration in bandit problems with side observations
Dec 5th Jean-Bastien Grill Hierarchical optimistic optimization for X-armed bandit
Dec 18th Lihong Li Multi-armed Bandits on the Internet salle plenière
Dec 19th Lihong Li On Minimax Optimal Offline Policy Evaluation

2014 spring/summer:

Date Speaker Title
Jan 10th Prashanth L. A. Actor-critic algorithms for risk-sensitive MDPs
Feb 14th Sergio Valcarcel Diffusion strategies for cooperative reinforcement learning
Feb 28th Alessandro Lazaric Regret Bounds for Reinforcement Learning with Policy Advice
Mar 12th Matthieu Geist Around Inverse Reinforcement Learning and Score-based Classification
Mar 14th Bilal Piot Reinforcement Learning with Expert Demonstrations and links with IRL
Apr 28th Jennifer Healey Transportation Futures: Gossiping Cars and Chatty Cities
May 16th Tomáš Kocák Spectral Bandits for Smooth Graph Functions
May 20th Pascal Poupart Online Bayesian Moment Matching for Latent Dirichlet Allocation
May 23rd Djalel Benbouzid Sequential budgeted classification with complex cost-dependency structure
Jun 13th Mylene Maida An overview of some recent results on deformed random matrix models
Jul 16th Marta Soare Best-Arm Identification in Linear Bandits
Sep 5th Prashanth L. A. Stochastic approximation for speeding up LSTD (and LSPI)

Seminars from even longer ago

Comments are closed