Multi-scale exploration of convex functions and bandit convex optimization

@inproceedings{Bubeck2016MultiscaleEO,
  title={Multi-scale exploration of convex functions and bandit convex optimization},
  author={S{\'e}bastien Bubeck and Ronen Eldan},
  booktitle={COLT},
  year={2016}
}
We construct a new map from a convex function to a distributio n on its domain, with the property that this distribution is a multi-scale explorati on of the function. We use this map to solve a decade-old open problem in adversarial bandit conve x optimization by showing that the minimax regret for this problem is̃ O(poly(n) √ T ), wheren is the dimension andT the number of rounds. This bound is obtained by studying the dual Bayesian maximin regret via the information ratio analysis of Russo… CONTINUE READING
Highly Cited
This paper has 20 citations. REVIEW CITATIONS
Related Discussions
This paper has been referenced on Twitter 6 times. VIEW TWEETS

From This Paper

Topics from this paper.

References

Publications referenced by this paper.
Showing 1-5 of 5 references

Bandit convex opt imization

Bubeck, O. Dekel, T. Koren, Y. Peres
Annual Conference on Learning Theor y (COLT), • 2012

Regret analysis of stochastic and nonstochastic multiarmed bandit problems

S. Bubeck, N. Cesa-Bianchi
Foundations and Trends in Machine Learning • 2012

Multi - scale exploration of convex functions and bandit convex optimization

S. Bubeck, R. Eldan
Arxiv preprint arXiv

Similar Papers

Loading similar papers…