Institut de Mathématiques de Toulouse

Les événements de la journée


1 événement


  • Mathématiques de l’apprentissage

    Lundi 3 septembre 12:00-13:00 - Andrea Locatelli - Universität Magdeburg

    Adaptivity to Smoothness in X-armed bandits

    Résumé : We study the stochastic continuum-armed bandit problem from the angle of adaptivity to unknown regularity of the reward function f. We prove that there exists no strategy for the cumulative regret that adapts optimally to the smoothness of f. We show however that such minimax optimal adaptive strategies exist if the learner is given extra-information about f. Finally, we complement our positive results with matching lower bounds. The paper was published at COLT 2018 (link to the paper : http://proceedings.mlr.press/v75/locatelli18a.html).

    Lieu : Bât 1R3, salle de conférence MIP.

    [En savoir plus]