CACTO: Continuous Actor-Critic with Trajectory Optimization - Towards global optimality.
Gianluigi GrandessoGastone P. Rosati PapiniPatrick M. WensingAndrea Del PretePublished in: CoRR (2022)
Keyphrases
- global optimality
- global optimization
- globally optimal
- discrete optimization
- actor critic
- semidefinite
- optimal solution
- theoretical guarantees
- reinforcement learning
- global minimum
- optimal control
- optimization algorithm
- objective function
- optimization problems
- convex programming
- evaluation function
- optimization methods
- sufficient conditions
- temporal difference
- particle swarm optimization
- linear programming
- evolutionary algorithm
- gradient method
- policy gradient
- machine learning