Composing and combining policies under the policy machine.
David F. FerraioloSerban I. GavrilaVincent C. HuD. Richard KuhnPublished in: SACMAT (2005)
Keyphrases
- optimal policy
- policy search
- management policies
- access control policies
- allocation policies
- allocation policy
- control policies
- transport systems
- policy gradient methods
- revenue management
- markov decision process
- reinforcement learning
- control policy
- reward function
- infinite horizon
- dynamic programming
- access control
- state dependent
- sufficient conditions
- expected reward
- decision processes
- batch processing
- neural network
- selective perception
- state space
- optimal pricing
- markov decision processes
- production rate
- decision problems
- markov decision problems
- security policies
- conflict resolution
- privacy policies
- finite horizon
- combining multiple
- asymptotically optimal
- long run
- average cost