Finding the maximum with linear error probabilities: a sequential analysis approach
Assume that n players are represented by n reals, uniformly distributed over the unit interval.
We assume that the error probability of a comparison game between two players depends linearly on the distance between the players. Using sequential analysis approach, we present an algorithm to estimate the maximum ξ of the players with an error less than ε.
Mean cost, variance and centered moments generating function are analyzed.
Unable to display preview. Download preview PDF.
- 1.M. Adler, P. Gemmell, M. Harchol, R.M. Karp and Cl. Kenyon Selection in the Presence of Noise: The Design of Playoff Systems Proc. SODA 1994, San Francisco.Google Scholar
- 2.D.A. Darling and J.F. Siegert The First Passage Problem for a Continuous Markov Process An. Math. Stat. 1953, 624–639.Google Scholar
- 3.U. Feige, D. Peleg, P. Raghavan and E. Upfal Computing with Unreliable Information. In Symposium on Theory of Computing, 1990, 128–137.Google Scholar
- 4.B.K. Gosh and P.K. Sen Handbook of Sequential Analysis. Marcel Dekker, 1991.Google Scholar
- 5.G. Louchard, R. Schott Probabilistic Analysis of Some Distributed Algorithms. Wiley, Random Structures and Algorithms 2, 2, 1991, 151–186.Google Scholar
- 6.G. Louchard Finding the maximum with linear error probabilities: a sequential analysis approach, Département d'Informatique, TR-297, 1994.Google Scholar
- 7.V.V. Petrov Sums of Independent Random Variables Ergebnisse der Mathematik und ihrer Grenzgebiete. Band 82, Springer-Verlag, 1975.Google Scholar