OPTIMUM POLICIES FOR PARTIALLY OBSERVABLE MARKOV SYSTEMS
Abstract
The problem considered is the monitoring of a discrete-state Markov process through a noisy channel with associated costs. Dynamic programming techniques are used to establish the optimal monitoring decision policies for both transient and steady-state situations. Examples are included.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1965
- Accession Number
- AD0480213
Entities
People
- James S. Kakalik
Organizations
- Massachusetts Institute of Technology