Reinforcement Learning in Partially Observable Markov Decision Processes using Hybrid Probabilistic Logic Programs
Emad SaadPublished in: CoRR (2010)
Keyphrases
- partially observable markov decision processes
- reinforcement learning
- continuous state
- optimal policy
- partially observable domains
- policy search
- partial observability
- state space
- markov decision processes
- function approximation
- finite state
- belief space
- hidden state
- partially observable environments
- partially observable stochastic games
- policy gradient
- partially observable
- model free
- reinforcement learning algorithms
- belief state
- stochastic domains
- dynamic programming
- temporal difference
- decision problems
- multi agent
- machine learning
- dynamical systems
- partially observable markov decision process
- reward function
- markov decision problems
- logic programs
- fully observable