Boltzmann Exploration Done Right

@inproceedings{CesaBianchi2017BoltzmannED,
  title={Boltzmann Exploration Done Right},
  author={Nicol{\`o} Cesa-Bianchi and Claudio Gentile and G{\'a}bor Lugosi and Gergely Neu},
  booktitle={NIPS},
  year={2017}
}
Boltzmann exploration is a classic strategy for sequential decision-making under uncertainty, and is one of the most standard tools in Reinforcement Learning (RL). Despite its widespread use, there is virtually no theoretical understanding about the limitations or the actual benefits of this exploration scheme. Does it drive exploration in a meaningful way? Is it prone to misidentifying the optimal actions or spending too much time exploring the suboptimal ones? What is the right tuning for the… CONTINUE READING

From This Paper

Figures and tables from this paper.

References

Publications referenced by this paper.
Showing 1-10 of 26 references

Challenging the empirical mean and empirical variance: A deviation study

O. Catoni
Annales de l’Institut Henri Poincaré, Probabilités et Statistiques, 48(4):1148–1185, • 2012
View 4 Excerpts
Highly Influenced

Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems

S. Bubeck, N. Cesa-Bianchi
Now Publishers Inc, • 2012
View 1 Excerpt

Similar Papers

Loading similar papers…