Deep Reinforcement Learning in Parameterized Action Space.
Matthew J. HausknechtPeter StonePublished in: ICLR (Poster) (2016)
Keyphrases
- action space
- reinforcement learning
- state space
- markov decision processes
- state and action spaces
- continuous state
- real valued
- control policies
- reinforcement learning methods
- continuous state spaces
- function approximators
- action selection
- state action
- function approximation
- multi agent
- single agent
- stochastic processes
- markov decision process
- learning algorithm
- dynamic programming
- reinforcement learning algorithms
- optimal policy
- markov decision problems
- cooperative
- markov chain
- control problems
- temporal difference
- planning problems