Policy ensemble gradient for continuous control problems in deep reinforcement learning.
Guoqiang LiuGang ChenVictoria HuangPublished in: Neurocomputing (2023)
Keyphrases
- control problems
- reinforcement learning
- continuous state spaces
- policy gradient
- action space
- optimal policy
- optimal control
- policy search
- state space
- reinforcement learning algorithms
- stochastic control
- reinforcement learning methods
- function approximation
- markov decision process
- action selection
- markov decision problems
- actor critic
- queueing systems
- partially observable
- learning algorithm
- partially observable markov decision processes
- state action
- adaptive control
- rl algorithms
- continuous state
- approximate dynamic programming
- policy iteration
- multi agent
- function approximators
- markov decision processes
- control policies
- dynamic programming
- control policy
- model free
- data mining
- markov games
- gradient method
- reward function
- decision problems
- markov chain
- machine learning