Autonomous agent learning using an actor-critic algorithm and behavior models.
Víctor Uc CetinaPublished in: AAMAS (3) (2008)
Keyphrases
- actor critic
- learning algorithm
- reinforcement learning
- dynamic programming
- policy gradient
- optimal solution
- monte carlo
- neuro fuzzy
- search space
- autonomous agents
- learning problems
- approximate dynamic programming
- linear programming
- dynamic environments
- convergence rate
- function approximation
- optimal control
- reinforcement learning algorithms
- multi agent