Expected Policy Gradients for Reinforcement Learning.
Kamil CiosekShimon WhitesonPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- optimal policy
- total reward
- policy search
- markov decision process
- action selection
- markov decision processes
- reinforcement learning algorithms
- partially observable environments
- reward function
- state space
- function approximation
- function approximators
- reinforcement learning problems
- control policy
- partially observable
- policy gradient
- state and action spaces
- learning algorithm
- action space
- control policies
- actor critic
- markov decision problems
- partially observable domains
- partially observable markov decision processes
- multi agent
- opportunity cost
- rl algorithms
- finite state
- dynamic programming
- policy iteration
- machine learning
- infinite horizon
- long run
- stationary policies
- robotic control
- continuous state spaces
- agent learns
- neural network
- sufficient conditions
- decision problems
- transition model
- optimal control
- policy evaluation
- average reward
- model free
- temporal difference