Dynamic Programming for One-sided Partially Observable Pursuit-evasion Games.
Karel HorákBranislav BosanskýPublished in: ICAART (2) (2017)
Keyphrases
- partially observable
- pursuit evasion
- dynamic programming
- state space
- infinite horizon
- markov decision processes
- markov decision problems
- reinforcement learning
- decision problems
- optimal policy
- partial observability
- dynamical systems
- optimal control
- game theory
- dec pomdps
- partially observable domains
- partial observations
- reward function
- action models
- belief state
- heuristic search
- partially observable environments
- orders of magnitude
- average cost
- partially observable markov decision processes
- linear programming
- nash equilibrium
- state variables
- finite state
- policy iteration
- multi agent
- computational complexity
- learning algorithm
- search space
- markov decision process
- stereo matching