Reinforcement Learning Using Monte Carlo Policy Estimation for Disaster Mitigation.
Mohammed Talat KhoujSarbjit SarkariaCesar Lopez CastellanosJosé R. MartíPublished in: Critical Infrastructure Protection (2014)
Keyphrases
- monte carlo
- policy evaluation
- reinforcement learning
- importance sampling
- monte carlo simulation
- temporal difference
- optimal policy
- stochastic approximation
- markov chain
- policy search
- markov decision process
- monte carlo methods
- matrix inversion
- policy iteration
- action selection
- simulation study
- markovian decision
- function approximators
- monte carlo tree search
- temporal difference learning
- particle filter
- monte carlo method
- action space
- semi parametric
- point processes
- learning algorithm
- finite state
- function approximation
- markov decision processes
- machine learning
- model free
- adaptive sampling
- state space
- reinforcement learning algorithms
- variance reduction
- markov chain monte carlo
- optimal strategy
- global illumination
- policy gradient
- confidence intervals
- parameter estimation
- decision support system