An Improved Parametrization and Analysis of the EXP3++ Algorithm for Stochastic and Adversarial Bandits

Abstract

We present a new strategy for gap estimation in randomized algorithms for multiarmed bandits and combine it with the EXP3++ algorithm of Seldin and Slivkins (2014). In the stochastic regime the strategy reduces dependence of regret on a time horizon from $(\ln t)^3$ to $(\ln t)^2$ and eliminates an additive factor of order $∆e^1/∆^2$, where $∆$ is the minimal gap of a problem instance. In the adversarial regime regret guarantee remains unchanged.

Cite

Text

Seldin and Lugosi. "An Improved Parametrization and Analysis of the EXP3++ Algorithm for Stochastic and Adversarial Bandits." Proceedings of the 2017 Conference on Learning Theory, 2017.

Markdown

[Seldin and Lugosi. "An Improved Parametrization and Analysis of the EXP3++ Algorithm for Stochastic and Adversarial Bandits." Proceedings of the 2017 Conference on Learning Theory, 2017.](https://mlanthology.org/colt/2017/seldin2017colt-improved/)

BibTeX

@inproceedings{seldin2017colt-improved,
  title     = {{An Improved Parametrization and Analysis of the EXP3++ Algorithm for Stochastic and Adversarial Bandits}},
  author    = {Seldin, Yevgeny and Lugosi, Gábor},
  booktitle = {Proceedings of the 2017 Conference on Learning Theory},
  year      = {2017},
  pages     = {1743-1759},
  volume    = {65},
  url       = {https://mlanthology.org/colt/2017/seldin2017colt-improved/}
}