General policies for large weakly-coupled multi-action restless bandits converge exponentially in probability to a deterministic limit, and a proposed policy achieves exponential asymptotic optimality as system size increases.
Weakly coupled dyna mic program: Information and Lagrangian relaxations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Weakly-Coupled Multi-Action Restless Bandits -- Exponential Convergence in Probability
General policies for large weakly-coupled multi-action restless bandits converge exponentially in probability to a deterministic limit, and a proposed policy achieves exponential asymptotic optimality as system size increases.