Human feedback in continuous actor-critic reinforcement learning.
Cristian MillánBruno J. T. FernandesFrancisco CruzPublished in: ESANN (2019)
Keyphrases
- cost function
- actor critic
- reinforcement learning
- temporal difference
- policy gradient
- reinforcement learning algorithms
- optimal control
- approximate dynamic programming
- function approximation
- neuro fuzzy
- policy iteration
- action space
- markov decision processes
- gradient method
- model free
- state space
- objective function
- optimal policy
- function approximators
- temporal difference learning
- policy gradient methods
- control problems
- learning problems
- partially observable
- action selection
- infinite horizon
- negative matrix factorization
- evaluation function
- real valued
- reinforcement learning methods
- fixed point
- linear programming
- fuzzy logic