Prediction with Expert Advice by Following the Perturbed Leader for General Weights

@inproceedings{Hutter2004PredictionWE,
  title={Prediction with Expert Advice by Following the Perturbed Leader for General Weights},
  author={Marcus Hutter and Jan Poland},
  booktitle={ALT},
  year={2004}
}
When applying aggregating strategies to Prediction with Expert Advice, the learning rate must be adaptively tuned. The natural choice of √ complexity/current loss renders the analysis of Weighted Majority derivatives quite complicated. In particular, for arbitrary weights there have been no results proven so far. The analysis of the alternative “Follow the Perturbed Leader” (FPL) algorithm from [KV03] (based on Hannan’s algorithm) is easier. We derive loss bounds for adaptive learning rate and… CONTINUE READING

From This Paper

Topics from this paper.
20 Citations
13 References
Similar Papers

References

Publications referenced by this paper.
Showing 1-10 of 13 references

In Proceedings of the 16th Annual Conference on Learning Theory (COLT-2003)

  • A. Kalai, S. Vempala. Efficient algorithms for online decision
  • Lecture Notes in Artificial Intelligence, pages…
  • 2003
Highly Influential
6 Excerpts

Journal of the ACM

  • N. Cesa-Bianchi et al. How to use expert advice
  • 44(3):427–485,
  • 1997
2 Excerpts

pages 371–383

  • V. G. Vovk. Aggregating strategies. In Proceedings of Theory
  • Rochester, New York,
  • 1990
1 Excerpt

Similar Papers

Loading similar papers…