Journals

J.-Y. Audibert, R. Munos, Cs. Szepesvári, Exploration-exploitation trade-off using variance estimates in multi-armed bandits, Theoretical Computer Science, Vol. 410, p.1876-1902, 2009

J.-Y. Audibert, Fast learning rates in statistical inference through aggregation, Annals of statistics, a note on the VC lower bound, Vol. 37, p.1591-1646, 2009

J.-Y. Audibert, O. Bousquet, Combining PAC-Bayesian and generic chaining bounds, Journal of Machine Learning Research, Vol. 8, p.863-889, 2007

J.-Y. Audibert, A.Tsybakov, Fast learning rates for plug-in classifiers, Long preprint version here, Annals of statistics, Vol. 35, No. 2, p.608-633, April 2007. Slides for the Empirical Processes and Asymptotic Statistics Workshop, 2007

J.-Y. Audibert, Aggregated estimators and empirical complexity for least square regression, Preprint version, Annales de l'Institut Henri Poincare (B) Probability and Statistics, Volume 40, Issue 6, p.685-736, November-December 2004

PhD thesis

J.-Y. Audibert, Théorie Statistique de l’Apprentissage : une approche PAC-Bayésienne, thčse de doctorat de l’Université Paris VI, June 2004, slides

Conferences

J.-Y. Audibert, S. Bubeck, Minimax policies for adversarial and stochastic bandits, 22nd Annual Conference on Learning Theory (COLT), 2009

Y. Wang, J.-Y. Audibert, R. Munos, Algorithms for Infinitely Many-Armed Bandits, Supplemental material, Advances in Neural Information Processing Systems 21 (NIPS), 2008

V. Mnih, Cs. Szepesvari, J.-Y. Audibert, Empirical Bernstein stopping, 25th International Conference on Machine Learning (ICML), 2008

J.-Y. Audibert, Progressive mixture rules are deviation suboptimal, Advances in Neural Information Processing Systems 20 (NIPS), 2007, Supplemental material (1,2)

J.-Y. Audibert, R. Munos, Cs. Szepesvári, Tuning Bandit Algorithms in Stochastic Environments, 18th International Conference on Algorithmic Learning Theory (ALT), 2007

J.-Y. Audibert, R. Munos, Cs. Szepesvári, NIPS 2006 Workshop on On-line Trading of Exploration and Exploitation, Use of variance estimation in the multi-armed bandit problem, slides here, 2006

J.-Y. Audibert, A randomized online learning algorithm for better variance control, slides here, long version here. 19th Annual Conference on Learning Theory (COLT), p.392-407, 2006

Technical reports

J.-Y. Audibert and O. Catoni, Risk bounds in linear regression through PAC-Bayesian truncation, some related slides (v1,v2) and some introductory slides to the PAC-Bayes bounds and for reading the first chapter of Olivier Catoni's book, 2009

J.-Y. Audibert, No fast exponential deviation inequalities for the progressive mixture rule, CERTIS Technical report 07-35, 2007

J.-Y. Audibert, R. Munos, Cs. Szepesvári, Variance estimates and exploration function in multi-armed bandit, CERTIS Technical report 07-31, 2007

J.-Y. Audibert, Fast learning rates in statistical inference through aggregation, long version of the one in the Annals of Statistics, CERTIS Technical report 06-20, 2006