Continuous Value Iteration (CVI) Reinforcement Learning and Imaginary Experience Replay (IER) for learning multi-goal, continuous action and state space controllers.
Andreas GerkenMichael SprangerPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- state space
- action space
- markov decision processes
- continuous state spaces
- policy search
- continuous action
- continuous state
- learning algorithm
- learning problems
- continuous state and action spaces
- heuristic search
- optimal policy
- reinforcement learning algorithms
- model free
- partially observable markov decision processes
- function approximation
- optimal control
- supervised learning
- markov chain
- partially observable
- reinforcement learning methods
- multi agent
- reward signal
- dynamic programming
- particle filter
- markov decision process
- temporal difference
- learning tasks
- learning agent
- state transition
- state variables
- transfer learning