Preprints

• Optimal Multi-Fidelity Best Arm Identification. Riccardo Poiani, Rémy Degenne, Emilie Kaufmann, Alberto Maria Metelli, Marcello Restelli.
• Finding good policies in average-reward Markov Decision Processes without prior knowledge. . Adrienne Tuynman, Rémy Degenne, Emilie Kaufmann.
• Towards Instance-Optimality in Online PAC Reinforcement Learning. Aymen Al-Marjani, Andrea Tirinzoni and Emilie Kaufmann.

2024

• Power Mean Estimation in Stochastic Monte-Carlo Tree Search. Tuan Dam, Odalric-Ambrym Maillard, Emilie Kaufmann. Conference on Uncertainty in Artificial Intelligence (UAI).
• Bandit Pareto Set Identification: The Fixed Budget Setting. Cyrille Kone, Emilie Kaufmann and Laura Richert. 27th International Conference on Artificial Intelligence and Statistics (AISTATS).

2023

• Adaptive Algorithms for Relaxed Pareto Set Identification. Cyrille Kone, Emilie Kaufmann and Laura Richert. Advances in Neural Information Processing Systems (NeurIPS), also presented at EWRL 2023.
• An ε-Best-Arm Identification Algorithm for Fixed-Confidence and Beyond. Marc Jourdan, Rémy Degenne and Emilie Kaufmann. Advances in Neural Information Processing Systems (NeurIPS), also presented at EWRL 2023.
• Active Coverage for PAC Reinforcement Learning. Aymen Al-Marjani, Andrea Tirinzoni and Emilie Kaufmann. Proceedings of the 36th Conference On Learning Theory (COLT).
• Dealing with Unknown Variances in Best Arm Identification. Marc Jourdan, Rémy Degenne, Emilie Kaufmann. International Conference on Algorithmic Learning Theory (ALT).
• Optimistic PAC Reinforcement Learning: the Instance-Dependent View. Andrea Tirinzoni, Aymen Al-Marjani and Emilie Kaufmann. International Conference on Algorithmic Learning Theory (ALT).

2022

• Top Two algorithms revisited. Marc Jourdan, Rémy Degenne, Dorian Baudry, Rianne de Heide and Emilie Kaufmann. Advances in Neural Information Processing Systems (NeurIPS).
• Near Optimal Collaborative Learning in Bandits. Clémence Réda, Sattar Vakili and Emilie Kaufmann. Advances in Neural Information Processing Systems (NeurIPS).
• Near Instance-Optimal PAC Reinforcement Learning for Deterministic MDPs. Andrea Tirinzoni, Aymen Al-Marjani and Emilie Kaufmann. Advances in Neural Information Processing Systems (NeurIPS), European Workshop on Reinforcement Learning (EWRL).
• Optimistic PAC Reinforcement Learning: the Instance-Dependent View. Andrea Tirinzoni, Aymen Al-Marjani and Emilie Kaufmann. European Workshop on Reinforcement Learning (EWRL).
• Efficient Algorithms for Extreme Bandits. Dorian Baudry, Yoan Russac and Emilie Kaufmann. 25th International Conference on Artificial Intelligence and Statistics (AISTATS).
• Efficient Change-Point Detection for Tackling Piecewise-Stationary Bandits. Lilian Besson, Emilie Kaufmann, Odalric-Ambrym Maillard and Julien Seznec. Journal of Machine Learning Research, 23(77): 1-40.

2021

• Mixture Martingales Revisited with Applications to Sequential Tests and Confidence Intervals. Emilie Kaufmann and Wouter M. Koolen. Journal of Machine Learning Research, 22(246), 1-44.
• Optimal Thompson Sampling strategies for support-aware CVaR bandits. Dorian Baudry, Romain Gautron, Emilie Kaufmann and Odalric-Ambrym Maillard. International Conference on Machine Learning (ICML).
• Fast Active Learning for Pure Exploration in Reinforcement Learning . Pierre Ménard, Omar Darwiche Domingues, Anders Jonsson, Emilie Kaufmann, Edouard Leurent and Michal Valko. International Conference on Machine Learning (ICML).
• Kernel-Based Reinforcement Learning: a Finite-Time Analysis. Omar Darwiche Domingues, Pierre Ménard, Matteo Pirotta, Emilie Kaufmann and Michal Valko. International Conference on Machine Learning (ICML).
• A Kernel-Based Approach to Non-Stationary Reinforcement Learning in Metric Spaces. Omar Darwiche Domingues, Pierre Ménard, Matteo Pirotta, Emilie Kaufmann and Michal Valko. 24th International Conference on Artificial Intelligence and Statistics (AISTATS).
• Top-m identification for linear bandits. Clémence Réda, Emilie Kaufmann and Andrée Delahaye-Duriez. 24th International Conference on Artificial Intelligence and Statistics (AISTATS).
• Non-Asymptotic Sequential Tests for Overlapping Hypotheses and application to near optimal arm identification in bandit models. Aurélien Garivier and Emilie Kaufmann. Sequential Analysis, 40(1):61-96. Julia code for best arm identification.
• Episodic Reinforcement Learning in Finite MDPs: Minimax Lower Bounds Revisited . Omar Darwiche Domingues, Pierre Ménard, Emilie Kaufmann and Michal Valko. International Conference on Algorithmic Learning Theory (ALT).
• Adaptive Reward-Free Exploration. Emilie Kaufmann, Pierre Ménard, Omar Darwiche Domingues, Anders Jonsson, Edouard Leurent and Michal Valko. International Conference on Algorithmic Learning Theory (ALT).
• On Multi-Armed Bandit Designs for Dose-Finding Trials. Maryam Aziz, Emilie Kaufmann and Marie-Karelle Rivière. Journal of Machine Learning Research, 22(14):1−38.

2020

• Sub-sampling for Efficient Non-Parametric Bandit Exploration. Dorian Baudry, Emilie Kaufmann and Odalric-Ambrym Maillard. Advances in Neural Information Processing Systems (NeurIPS).
• Planning in Markov Decision Processes with Gap-Dependent Sample Complexity. Anders Jonsson, Emilie Kaufmann, Pierre Ménard, Omar Darwiche Domingues, Edouard Leurent and Michal Valko. Advances in Neural Information Processing Systems (NeurIPS).
• Adaptive Reward-Free Exploration. Emilie Kaufmann, Pierre Ménard, Omar Darwiche Domingues, Anders Jonsson, Edouard Leurent and Michal Valko. Theoretical Foundations of Reinforcement Learning Workshop @ICML. Video.
• A Kernel-Based Approach to Non-Stationary Reinforcement Learning in Metric Spaces. Omar Darwiche Domingues, Pierre Ménard, Matteo Pirotta, Emilie Kaufmann and Michal Valko. Theoretical Foundations of Reinforcement Learning Workshop @ICML. Video.
• Fixed Confidence Guarantees for Bayesian Best Arm Identification. Xuedong Shang, Rianne de Heide, Emilie Kaufmann, Pierre Ménard and Michal Valko. 23rd International Conference on Artificial Intelligence and Statistics (AISTATS).
• A Practical Algorithm for Multiplayer Bandits when Arm Means Vary Among Players. Etienne Boursier, Emilie Kaufmann, Abbas Mehrabian and Vianney Perchet. 23rd International Conference on Artificial Intelligence and Statistics (AISTATS).
• Solving Bernoulli Rank-One Bandits with Unimodal Thompson Sampling.
Cindy Trinh, Emilie Kaufmann, Claire Vernade and Richard Combes. International Conference on Algorithmic Learning Theory (ALT).
• Machine learning applications in drug development.
Clémence Réda, Emilie Kaufmann, Andrée Delahaye-Duriez. Computational and Structural Biotechnology Journal 18: 241-252.

2019

• Analyse non asymptotique d'un test séquentiel de détection de ruptures et application aux bandits non stationnaires(in French). Lilian Besson and Emilie Kaufmann. GRETSI.
• A simple dynamic bandit algorithm for hyper-parameter tuning.
Xuedong Shang, Emilie Kaufmann and Michal Valko. AutoML workshop @ ICML.
• Asymptotically Optimal Algorithms for Budgeted Multiple Play Bandits.
Alexander Luedtke, Emilie Kaufmann and Antoine Chambaz. Machine Learning 108(11): 1919-1949.
• General Parallel Optimization without Metric. Xuedong Shang, Emilie Kaufmann and Michal Valko. International Conference on Algorithmic Learning Theory (ALT).

2018

• Sequential Test for the Lowest Mean: From Thompson to Murphy Sampling. Emilie Kaufmann, Wouter M. Koolen and Aurélien Garivier. Advances in Neural Information Processing Systems (NeurIPS).
• Adaptive Black-Box Optimization Got Easier: HCT Only Needs Local Smoothness. Xuedong Shang, Emilie Kaufmann and Michal Valko. European Workshop on Reiforcement Learning (EWRL).
• Multi-Player Bandits Revisited. Lilian Besson and Emilie Kaufmann. International Conference on Algorithmic Learning Theory (ALT). Erratum for the Lower Bound.
• Corrupt Bandits for Preserving Local Privacy. Pratik Gajane, Tanguy Urvoy and Emilie Kaufmann. International Conference on Algorithmic Learning Theory (ALT).
• Pure Exploration in Infinite Bandit Models with Fixed Confidence. Maryam Aziz, Jesse Anderton, Emilie Kaufmann and Javed Aslam. International Conference on Algorithmic Learning Theory (ALT).
• Aggregation of Multi-Armed Bandits learning algorithms for Opportunistic Spectrum Access. Lilian Besson, Emilie Kaufmann, Christophe Moy. IEEE Wireless Communications and Networking Conference (WCNC).
  

2017

• Learning the distribution with largest mean: two bandit frameworks. Emilie Kaufmann and Aurélien Garivier. ESAIM: Proceedings and Surveys, Vol 60:114-131.
• Monte-Carlo Tree Search by Best Arm Identification. Emilie Kaufmann and Wouter M. Koolen. Advances in Neural Processing Systems (NIPS).
• Multi-Armed Bandit Learning in IoT Networks: Learning helps even in non-stationary settings. Rémi Bonnefoi, Lilian Besson, Christophe Moy, Emilie Kaufmann and Jacques Palicot. International Conference on Cognitive Radio Oriented Wireless Networks (CROWNCOM). (best paper award)
• A Spectral Algorithm with Additive Clustering for the Recovery of Overlapping Communities in Networks. Emilie Kaufmann, Thomas Bonald and Marc Lelarge. Journal of Theoretical Computer Science.
  
• On Bayesian Index Policies for Sequential Resource Allocation. Emilie Kaufmann. Annals of Statistics, Vol 46(2): 842-865.

2016

• Modèles de bandit : une histoire bayésienne et fréquentiste (survey paper in French). MATAPLI 109:51-64, 2016.
• On Explore-Then-Commit Strategies. Aurélien Garivier, Emilie Kaufmann and Tor Lattimore. Advances in Neural Information Processing Systems (NIPS).
• Corrupt Bandits. Pratik Gajane, Tanguy Urvoy and Emilie Kaufmann. European Workshop on Reinforcement Learning (EWRL).
• A Spectral Algorithm with Additive Clustering for the Recovery of Overlapping Communities in Networks.
Emilie Kaufmann, Thomas Bonald and Marc Lelarge. International Conference on Algorithmic Learning Theory (ALT). (slides)

• Maximin Action Identification: a New Bandit Framework for Games,
Aurélien Garivier, Emilie Kaufmann and Wouter M. Koolen. Proceedings of the 29th Conference on Learning Theory (COLT). (slides)
• Optimal Best-Arm Identification with Fixed Confidence,
Aurélien Garivier and Emilie Kaufmann. Proceedings of the 29th Conference on Learning Theory (COLT). (Julia code for best arm identification.
• On the Complexity of Best Arm Identification in Multi-Armed Bandit Models.
Emilie Kaufmann, Aurélien Garivier and Olivier Cappé. Journal of Machine Learning Research, Vol 17(1):1-42.

2015

• A novel spectral algorithm for the identification of overlapping communities, Workshop on Features and Structures (FEAST) @ ICML. (best poster award)

2014

• On the Complexity of A/B Testing. Emilie Kaufmann, Aurélien Garivier and Olivier Cappé. Proceedings of the 27th Conference on Learning Theory (COLT). (slides)

2013

• Thompson Sampling for one-dimensional exponential family bandits. Nathaniel Korda, Emilie Kaufmann and Rémi Munos. in Advances in Neural Information Processing Systems (NIPS).
• Information complexity in bandit subset selection. Emilie Kaufmann and Shivaram Kalyanakrishnan, Proceedings of the 26th Conference On Learning Theory (COLT).
• Information complexity in bandit subset selection. Journées Francophones sur la Planification, la Décision et l'Apprentissage pour la conduite de systèmes (JFPDA). (best poster award)

2012

• Thompson Sampling: an asymptotically optimal finite-time analysis. Emilie Kaufmann, Nathaniel Korda and Rémi Munos, International Conference on Algorithmic Learning Theory (ALT). (slides)
• On Bayesian Upper Confidence Bounds for Bandits Problems. Emilie Kaufmann, Olivier Cappé and Aurélien Garivier, Fifteenth International Conference on Artificial Intelligence and Statistics (AISTATS). (poster)

2011

• On the efficiency of Bayesian bandits algorithms from a frequentist point of view Workshop 'Bayesian optimization, experimental design, and bandits' @ NIPS. (poster)

Theses

• Contributions to the Optimal Solution of Several Bandits Problems, HDR thesis, defended November 13th, 2020. Université de Lille. (slides)
• Analysis of Bayesian and Frequentist strategies for Sequential Resource Allocation, PhD thesis, defended October 1st, 2014. PhD supervised by Olivier Cappé, Aurélien Garivier and Rémi Munos. This PhD thesis was awarded the Prix de thèse Jacques Neveu. (slides)