Deterministic Policy Optimization by Combining Pathwise and Score Function Estimators for Discrete Action Spaces.
Daniel LevyStefano ErmonPublished in: AAAI (2018)
Keyphrases
- action space
- score function
- continuous action
- continuous state spaces
- continuous state
- state space
- state and action spaces
- policy search
- real valued
- markov decision processes
- control policies
- reinforcement learning
- interval valued
- action selection
- stochastic processes
- optimal policy
- state dependent
- fuzzy numbers
- intuitionistic fuzzy
- partially observable markov decision processes
- markov decision process
- state action
- single agent
- decision problems
- probabilistic model
- evolutionary algorithm
- genetic algorithm