Universally Measurable Policies in Dynamic Programming.
Steven E. ShreveDimitri P. BertsekasPublished in: Math. Oper. Res. (1979)
Keyphrases
- dynamic programming
- optimal policy
- markov decision processes
- state space
- infinite horizon
- markov decision problems
- policy search
- optimal control
- multistage
- markov decision process
- linear programming
- greedy algorithm
- partially observable markov decision processes
- control system
- decision problems
- similarity measure
- dynamic programming algorithms
- reward function
- neural network
- control policies
- dp matching
- sequence alignment
- long run
- case study
- information retrieval
- data mining