Using continuous action spaces to solve discrete problems.
Hado van HasseltMarco A. WieringPublished in: IJCNN (2009)
Keyphrases
- action space
- continuous action
- continuous state spaces
- state space
- reinforcement learning
- real valued
- markov decision processes
- continuous variables
- continuous domains
- control policies
- continuous state
- policy search
- state and action spaces
- decision problems
- stochastic processes
- multi agent systems
- image segmentation