Adaptive game playing using multiplicative weights

  title={Adaptive game playing using multiplicative weights},
  author={Yoav Freund and Robert E. Schapire and Park Avenue and F. Park},
We present a simple algorithm for playing a repeated game. We show that a player using this algorithm suffers average loss that is guaranteed to come close to the minimum loss achievable by any fixed strategy. Our bounds are non-asymptotic and hold for any opponent. The algorithm, which uses the multiplicative-weight methods of Littlestone and Warmuth, is analyzed using the Kullback-Liebler divergence. This analysis yields a new, simple proof of the minmax theorem, as well as a provable method… CONTINUE READING
Highly Influential
This paper has highly influenced 41 other papers. REVIEW HIGHLY INFLUENTIAL CITATIONS
Highly Cited
This paper has 413 citations. REVIEW CITATIONS

From This Paper

Topics from this paper.


Publications citing this paper.
Showing 1-10 of 277 extracted citations

Generalized multiagent learning with performance bound

Autonomous Agents and Multi-Agent Systems • 2007
View 4 Excerpts
Highly Influenced

414 Citations

Citations per Year
Semantic Scholar estimates that this publication has 414 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-10 of 43 references

Probability inequalities for sums of bounded random variables

Wassily Hoeffding
Journal of the American Statistical Association, • 1963
View 5 Excerpts
Highly Influenced

A Game of Prediction with Expert Advice

J. Comput. Syst. Sci. • 1998
View 4 Excerpts
Highly Influenced

The Weighted Majority Algorithm

View 6 Excerpts
Highly Influenced

Foster and Rakesh V . Vohra . Asymptotic calibration

P. Dean
Operations Research • 1998

Helmbold , Robert E . Schapire , Yoram Singer , and Manfred K . Warmuth . On - line portfolio selection using multiplicative updates

P. David
Mathematical Finance • 1998

Schapire . Game theory , on - line prediction and boosting

Yoav Freund, E Robert
Biometrika • 1998

Foster and Rakesh V . Vohra . A randomization rule for selecting forecasts

P. Dean
Regret in the on - line decision problem • 1997

Warmuth . Additive versus exponentiated gradient updates for linear prediction

Jyrki Kivinen, K Manfred
Information and Computation • 1997

Similar Papers

Loading similar papers…