论文信息 - A short proof of the Gittins index theorem

A short proof of the Gittins index theorem

Provides a short and elementary proof of the Gittins index theorem for the multi-armed bandit problem, for the case where each bandit is modeled as a finite-state semi-Markov process. The author also indicates how this proof can be extended to the branching bandits and Klimov problems.<<ETX>>

J. Tsitsiklis

[1] G. Klimov. Time-Sharing Service Systems. I , 1975 .

[2] P. Whittle. Multi‐Armed Bandits and the Gittins Index , 1980 .

[3] P. Whittle. Arm-Acquiring Bandits , 1981 .

[4] P. Whittle. Sequential Project Selection (Multi-Armed Bandits) and the Gittins Index , 1982 .

[5] Jean Walrand,et al. Extensions of the multiarmed bandit problem: The discounted case , 1985 .

[6] J. Tsitsiklis. A lemma on the multiarmed bandit problem , 1986 .

[7] Guy Pujolle,et al. Introduction to queueing networks , 1987 .

[8] J. Ben Atkinson,et al. An Introduction to Queueing Networks , 1988 .

[9] Gideon Weiss,et al. Branching Bandit Processes , 1988, Probability in the Engineering and Informational Sciences.

[10] J. Walrand,et al. Interchange arguments in stochastic scheduling , 1989 .

[11] J. Bather,et al. Multi‐Armed Bandit Allocation Indices , 1990 .

[12] R. Weber. On the Gittins Index for Multiarmed Bandits , 1992 .

[13] Dimitris Bertsimas,et al. Conservation laws, extended polymatroids and multi-armed bandit problems: a unified approach to ind exable systems , 2011, IPCO.