| Article ID: | iaor19912104 |
| Country: | India |
| Volume: | 12 |
| Issue: | 2 |
| Start Page Number: | 307 |
| End Page Number: | 313 |
| Publication Date: | May 1991 |
| Journal: | Journal of Information & Optimization Sciences |
| Authors: | Khan Zulfiqar Ali |
This paper addresses the problem of multi-armed Bandit for which the state space is finite. It is well known that for finite state space the computation of the dynamic allocation indices can be handled by linear programming. Here, attempts are made to drive an optimal policy for the finite state multi-armed Bandit problem using linear programming.