Action Space Shaping in Deep Reinforcement Learning.
Anssi KanervistoChristian SchellerVille HautamäkiPublished in: CoRR (2020)
Keyphrases
- action space
- reinforcement learning
- state space
- reward shaping
- markov decision processes
- state and action spaces
- continuous state
- real valued
- continuous state spaces
- reinforcement learning methods
- action selection
- control policies
- state action
- stochastic processes
- single agent
- function approximators
- reinforcement learning algorithms
- function approximation
- particle filter
- multi agent
- model free
- machine learning
- markov decision process
- learning algorithm
- optimal control
- optimal policy
- non stationary
- markov decision problems
- mobile robot