Title :
A learning algorithm for the finite-time two-armed bandit problem
Author :
Sato, Mitsuhisa ; Abe, Kiyohiko ; Takeda, H.
Author_Institution :
Dept. of Electrical Engng., Tohoku Univ., Sendai, Japan
Abstract :
A simple algorithm for the finite-time two-armed bandit problem is proposed. In this algorithm, the whole process is divided into the first estimating process and the next controlling process. Efficient methods involving the use of approximation for computing the optimal length of the estimating process are provided.
Keywords :
game theory; learning systems; approximation; controlling process; estimating process; finite-time; learning algorithm; optimal length; two-armed bandit problem; Algorithm design and analysis; Approximation methods; Estimation; Nickel; Niobium; Process control; Skeleton;
Journal_Title :
Systems, Man and Cybernetics, IEEE Transactions on
DOI :
10.1109/TSMC.1984.6313253
