PLAS: Latent Action Space for Offline Reinforcement Learning.
Wenxuan ZhouSujay BajracharyaDavid HeldPublished in: CoRR (2020)
Keyphrases
- action space
- reinforcement learning
- state space
- state and action spaces
- markov decision processes
- real valued
- continuous state
- latent variables
- action selection
- reinforcement learning methods
- control policies
- state action
- stochastic processes
- continuous state spaces
- function approximators
- single agent
- control problems
- markov decision problems
- function approximation
- multiple agents
- markov decision process
- policy iteration
- markov chain
- learning algorithm
- cooperative
- partially observable markov decision processes
- model free
- state variables
- path planning
- heuristic search
- optimal policy