Policy Prediction Network: Model-Free Behavior Policy with Model-Based Learning in Continuous Action Space.
Zac WellmerJames T. KwokPublished in: ECML/PKDD (3) (2019)
Keyphrases
- model free
- action space
- reinforcement learning
- continuous state spaces
- rl algorithms
- reinforcement learning methods
- state space
- state action
- action selection
- policy iteration
- continuous state
- average reward
- function approximators
- policy evaluation
- learning algorithm
- reinforcement learning algorithms
- function approximation
- real valued
- control policies
- markov decision process
- markov decision processes
- policy search
- state and action spaces
- policy gradient
- optimal policy
- active learning
- partially observable markov decision processes
- learning process
- temporal difference
- partially observable
- markov decision problems
- learning tasks
- training data