Stationary multi-choice bandit problems.

DOI10.1016/S0165-1889(99)00064-0MaRDI QIDQ5958100zbMATH OpenFDO

Authors Dirk Bergemann, Juuso Välimäki

Publication date 3 March 2002

Published in Journal of Economic Dynamics and Control (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/S0165-1889(99)00064-0

Mathematics Subject Classification ID

Stochastic scheduling theory in operations research (90B36)

Review text

This note shows that the optimal choice of \(k\) simultaneous experiments in a stationary multi-armed bandit problem can be characterized in terms of the Gittins index of each arm. The index characterization remains equally valid after the introduction of switching costs.

Recommendations

• Switching Costs and the Gittins Index
• Generalized Bandit Problems
• Gittins Index for Simple Family of Markov Bandit Processes with Switching Cost and No Discounting
• On the Gittins index for multiarmed bandits
• Multi-armed bandits with switching penalties

Cites work

• Denumerable-Armed Bandits
• Extensions of the multiarmed bandit problem: The discounted case
• scientific article; zbMATH DE number 3889341 (Why is no real title available?)
• scientific article; zbMATH DE number 194374 (Why is no real title available?)
• Optimal Search for the Best Alternative
• Switching Costs and the Gittins Index

Cited in

(25)

• Optimal hysteresis for a class of deterministic deteriorating two-armed bandit problem with switching costs.
• Branching bandits: A sequential search process with correlated pay-offs.
• Optimal learning and experimentation in bandit problems.
• Hierarchical experimentation
• Keeping your options open
• Large firms and within firm occupational reallocation
• Decomposing risk in an exploitation-exploration problem with endogenous termination time
• Cooperation dynamics in repeated games of adverse selection
• Optimal search from multiple distributions with infinite horizon
• Inefficiency of sponsored research
• Undiscounted bandit games
• Learning from failures: optimal contracts for experimentation and production
• On games of strategic experimentation
• Strategic information exchange
• Strategic learning in teams
• Robustness of stochastic bandit policies
• Competitive problem solving and the optimal prize schemes
• Switching Costs and the Gittins Index
• Two-armed restless bandits with imperfect information: stochastic control and indexability
• Gittins Index for Simple Family of Markov Bandit Processes with Switching Cost and No Discounting
• Generalized Bandit Problems
• A central limit theorem, loss aversion and multi-armed bandits
• Common value experimentation
• Strategic experimentation with private payoffs
• Response adaptive designs that incorporate switching costs and constraints

This page was built for publication: Stationary multi-choice bandit problems.

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5958100)