Improving Stochastic Policy Gradients in Continuous Control with Deep Reinforcement Learning using the Beta Distribution.
Po-Wei ChouDaniel MaturanaSebastian A. SchererPublished in: ICML (2017)
Keyphrases
- control policies
- reinforcement learning
- continuous state spaces
- action space
- control problems
- control policy
- optimal policy
- stochastic control
- state space
- action selection
- optimal control
- beta distribution
- model free reinforcement learning
- policy search
- function approximation
- continuous state
- direct policy search
- machine learning
- markov decision process
- markov decision processes
- dynamic programming
- temporal difference
- state dependent
- learning algorithm
- policy gradient
- model free
- actor critic
- high dimensional
- computer vision