• Computer Science, Mathematics
  • Published in ALT 2018

Cleaning up the neighborhood: A full classification for adversarial partial monitoring

@inproceedings{Lattimore2018CleaningUT,
  title={Cleaning up the neighborhood: A full classification for adversarial partial monitoring},
  author={Tor Lattimore and Cs. Szepesvari},
  booktitle={ALT},
  year={2018}
}
Partial monitoring is a generalization of the well-known multi-armed bandit framework where the loss is not directly observed by the learner. We complete the classification of finite adversarial partial monitoring to include all games, solving an open problem posed by Bartok et al. [2014]. Along the way we simplify and improve existing algorithms and correct errors in previous analyses. Our second contribution is a new algorithm for the class of games studied by Bartok [2013] where we prove… CONTINUE READING
2
Twitter Mentions

Figures and Topics from this paper.

Citations

Publications citing this paper.
SHOWING 1-7 OF 7 CITATIONS

An Information-Theoretic Approach to Minimax Regret in Partial Monitoring

VIEW 9 EXCERPTS
CITES BACKGROUND & METHODS

Exploration by Optimisation in Partial Monitoring

VIEW 17 EXCERPTS
CITES RESULTS & BACKGROUND

Partial monitoring

VIEW 7 EXCERPTS
CITES BACKGROUND & METHODS

Finding Robust Nash equilibria

VIEW 1 EXCERPT
CITES BACKGROUND

Connections Between Mirror Descent, Thompson Sampling and the Information Ratio

VIEW 1 EXCERPT
CITES BACKGROUND

On Adaptivity in Information-constrained Online Learning

VIEW 1 EXCERPT
CITES BACKGROUND

Stochastic Online Learning with Probabilistic Graph Feedback

VIEW 2 EXCERPTS
CITES BACKGROUND

References

Publications referenced by this paper.
SHOWING 1-6 OF 6 REFERENCES

On-Line Learning with Imperfect Monitoring

VIEW 5 EXCERPTS
HIGHLY INFLUENTIAL

The nonstochastic

  • Peter Auer, Nicolo Cesa-Bianchi, Yoav Freund, Robert E Schapire
  • Annual Symposium on,
  • 1995
VIEW 6 EXCERPTS
HIGHLY INFLUENTIAL

Generic Bound for Exponential Weights The proof of Theorem 2 depends on a generic regret analysis for a variant of the EXP3.P bandit algorithm by Auer et al

  • 1995

Probabilistic computations: Toward a unified measure of complexity

  • Andrew Chi-Chih Yao
  • Computer Science
  • 18th Annual Symposium on Foundations of Computer Science (sfcs 1977)
  • 1977