Near-Optimal No-Regret Algorithms for Zero-Sum Games

@inproceedings{Daskalakis2011NearOptimalNA,
  title={Near-Optimal No-Regret Algorithms for Zero-Sum Games},
  author={Constantinos Daskalakis and Alan Deckelbaum and Anthony Kim},
  booktitle={SODA},
  year={2011}
}
We propose a new no-regret learning algorithm. When used against an adversary, our algorithm achieves average regret that scales as O (1/√T) with the number T of rounds. This regret bound is optimal but not rare, as there are a multitude of learning algorithms with this regret guarantee. However, when our algorithm is used by both players of a zero-sum game, their average regret scales as O (ln T/T), guaranteeing a near-linear rate of convergence to the value of the game. This represents an… CONTINUE READING
Highly Cited
This paper has 41 citations. REVIEW CITATIONS

From This Paper

Topics from this paper.

References

Publications referenced by this paper.
Showing 1-7 of 7 references

Excessive Gap Technique in Nonsmooth Convex Minimization

SIAM Journal on Optimization • 2005
View 7 Excerpts
Highly Influenced

Smooth minimization of non-smooth functions

Math. Program. • 2005
View 4 Excerpts
Highly Influenced

An Iterative Method of Solving a Game

J. Robinson
Annals of Mathematics, 54:296– 301, • 1951
View 6 Excerpts
Highly Influenced

Theory of Games and Economic Behavior

Nature • 1946
View 4 Excerpts
Highly Influenced

Similar Papers

Loading similar papers…