A structured multiarmed bandit problem and the greedy policy.

Adam J. Mersereau Paat Rusmevichientong John N. Tsitsiklis

Published in: CDC (2008)

Keyphrases

multiarmed bandit
optimal policy
search algorithm
greedy algorithm
dynamic programming

structured data
greedy algorithms
greedy heuristic
feature selection
special case

genetic algorithm
active learning
decision process
asymptotically optimal
objective function

decision making
structured learning
policy making
policy search