Two-Target Algorithms for Infinite-Armed Bandits with Bernoulli Rewards

@inproceedings{Bonald2013TwoTargetAF,
  title={Two-Target Algorithms for Infinite-Armed Bandits with Bernoulli Rewards},
  author={Thomas Bonald and Alexandre Prouti{\`e}re},
  booktitle={NIPS},
  year={2013}
}
We consider an infinite-armed bandit problem with Bernoulli rewards. The mean rewards are independent, uniformly distributed over [0, 1]. Rewards 0 and 1 are referred to as a success and a failure, respectively. We propose a novel algorithm where the decision to exploit any arm is based on two successive targets, namely, the total number of successes until the first failure and until the first m failures, respectively, where m is a fixed parameter. This two-target algorithm achieves a long-term… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.

References

Publications referenced by this paper.
SHOWING 1-10 OF 14 REFERENCES

Optimal bayesian strategies for the infinite-armed bernoulli bandit

  • Ying-Chao Hung
  • Journal of Statistical Planning and Inference,
  • 2012
1 Excerpt

A note on infinite-armed bernoulli bandit problems with generalized beta prior distributions

  • Kung-Yu Chen, Chien-Tai Lin
  • Statistical Papers,
  • 2005
1 Excerpt

A note on strategies for bandit problems with infinitely many arms

  • Kung-Yu Chen, Chien-Tai Lin
  • Metrika
  • 2004

A note on strategies for bandit problems with infinitely many arms. Metrika

  • Kung-Yu Chen, Chien-Tai Lin
  • 2004
1 Excerpt

Some optimal strategies for bandit problems with beta prior distributions

  • Chien-Tai Lin, CJ Shiau
  • Annals of the Institute of Statistical…
  • 2000
2 Excerpts

Similar Papers

Loading similar papers…