Supervised Policy Update for Deep Reinforcement Learning.
Quan Ho VuongYiming ZhangKeith W. RossPublished in: ICLR (Poster) (2019)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- markov decision process
- learning algorithm
- action selection
- supervised learning
- state and action spaces
- reinforcement learning problems
- policy evaluation
- function approximators
- function approximation
- policy gradient
- state space
- partially observable environments
- reinforcement learning algorithms
- control policies
- control policy
- action space
- approximate dynamic programming
- markov decision processes
- temporal difference
- actor critic
- unsupervised learning
- semi supervised
- machine learning
- reward function
- partially observable
- decision problems
- policy iteration
- partially observable domains
- markov decision problems
- inverse reinforcement learning
- multi agent
- model free reinforcement learning
- continuous state
- state action
- rl algorithms
- infinite horizon
- dynamic programming
- training data
- transition model
- feature selection
- continuous state spaces
- partially observable markov decision processes
- agent learns
- transfer learning
- supervised classification
- optimal control
- temporal difference learning
- policy gradient methods
- learning process