Fuel in Markov Decision Processes (FiMDP): A Practical Approach to Consumption.
Frantisek BlahoudekMurat CubuktepePetr NovotnýMelkior OrnikPranay ThangedaUfuk TopcuPublished in: FM (2021)
Keyphrases
- markov decision processes
- optimal policy
- state space
- reinforcement learning
- transition matrices
- policy iteration
- finite state
- dynamic programming
- partially observable
- planning under uncertainty
- reachability analysis
- finite horizon
- decision processes
- risk sensitive
- average cost
- state and action spaces
- factored mdps
- infinite horizon
- decision theoretic planning
- reinforcement learning algorithms
- action sets
- state abstraction
- average reward
- mathematical model
- model based reinforcement learning
- dynamical systems
- data mining
- policy evaluation
- reward function
- multistage