论文信息 - Discrete multi-armed bandits and multi-parameter processes

Discrete multi-armed bandits and multi-parameter processes

SummaryThe general multi-armed bandit problem is reformulated and solved as a control problem over a partially ordered set. The approach taken provides a technically convenient framework for bandit-like problems. It also adds insight to the structure of strategies over partially ordered sets.

A. Mandelbaum

[1] J. Snell. Applications of martingale system theorems , 1952 .

[2] J. Neveu,et al. Discrete Parameter Martingales , 1975 .

[3] John B. Walsh,et al. Stochastic integrals in the plane , 1975 .

[4] Alan S. Willsky,et al. Optional Sampling of Submartingales Indexed by Partially Ordered Sets, Optional Sampling of Submartingales. , 1979 .

[5] J. Gittins. Bandit processes and dynamic allocation indices , 1979 .

[6] Stopping rules and tactics for processes indexed by a directed set , 1981 .

[7] Robert J. Vanderbei,et al. Optimal stopping and supermartingales over partially ordered sets , 1981 .

[8] Peter Whittle,et al. Optimization Over Time , 1982 .

[9] Mathematical learning models--theory and algorithms : proceedings of a conference , 1983 .

[10] R. Hartley,et al. Optimisation Over Time: Dynamic Programming and Stochastic Control: , 1983 .

[11] Markov Strategies for Optimal Control Problems Indexed by a Partially Ordered Set , 1983 .

[12] Arrêt Optimal sur le Plan , 1983 .

[13] P. Varaiya,et al. Extension of the multi-armed bandit problem , 1983, The 22nd IEEE Conference on Decision and Control.

[14] I. Karatzas. Gittins Indices in the Dynamic Allocation Problem for Diffusion Processes , 1984 .

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