Research in Adaptive and Decentralized Stochastic Control.
Interim rept. 15 Mar 84-14 Mar 85,
TEXAS UNIV AT AUSTIN DEPT OF ELECTRICAL AND COMPUTER ENGINEERING
Pagination or Media Count:
Significant progress was made in a number of aspects of stochastic systems. The problem of adaptive control of priority assignment in queueing systems was solved. A distance-measures approach to the problem of approximation and identification of queueing systems was studied. A problem of adaptively controlling a discounted-reward finite-state Markov decision process was solved. Major new results were obtained for the problem of adaptive control with incomplete observations. In particular, we have studied in depth a problem of adaptive control with incomplete observations, in which the state is a finite state Markov process. Author
- Statistics and Probability