Interests

Here you can find a (non-exhaustive) list of my research interests in random order:
  • Online Learning
  • Adaptive Methods in Online Learning
  • Adaptive Methods for Bandit Problems

Publications

Nonstochastic Bandits and Experts with Arm-Dependent Delays.
D. van der Hoeven and Nicolò Cesa-Bianchi. Accepted (oral) to AISTATS 2022.
Beyond Bandit Feedback in Online Multiclass Classification.
D. van der Hoeven, F. Fusco, and Nicolò Cesa-Bianchi. Advances in Neural Information Processing Systems, 2021
Sweater and Mug.
MetaGrad: Adaptation using Multiple Learning Rates in Online Learning.
T. van Erven, W.M. Koolen, and D. van der Hoeven. 2021. Journal of Machine Learning Research 22(161) pp. 1-61
Distributed Online Learning for Joint Regret with Communication Constraints.
D. van der Hoeven, Hédi Hadiji, and T. van Erven. accepted to ALT 2022.
Exploiting the Surrogate Gap in Online Multiclass Classification.
D. van der Hoeven. Advances in Neural Information Processing Systems 2020.
Poster. Presentation. Mug.
Comparator Adaptive Convex Bandits.
D. van der Hoeven, A. Cutkosky, and H. Luo. Advances in Neural Information Processing Systems 2020.
Poster.
Open Problem: Fast and optimal portfolio selection.
T. van Erven, D. van der Hoeven, W. Kotłowski, W.M. Koolen. Proceedings of the 33rd Conference on Learning Theory (COLT), pp. 3864-3869, 2020.
User-Specified Local Differential Privacy in Unconstrained Adaptive Online Learning
D. van der Hoeven. Advances in Neural Information Processing Systems, 2019. Poster.
The Many Faces of Exponential Weights in Online Learning
D. van der Hoeven, T. van Erven and W. Kotłowski. Proceedings of the 31st Conference on Learning Theory (COLT), pp. 2067-2092, 2018. Poster. Short presentation. Long presentation.