CACTO: Continuous Actor-Critic With Trajectory Optimization - Towards Global Optimality.
Gianluigi GrandessoElisa AlboniGastone Pietro Rosati PapiniPatrick M. WensingAndrea Del PretePublished in: IEEE Robotics Autom. Lett. (2023)
Keyphrases
- global optimality
- global optimization
- globally optimal
- discrete optimization
- actor critic
- optimal solution
- semidefinite
- reinforcement learning
- gradient method
- global minimum
- objective function
- optimization algorithm
- theoretical guarantees
- graph cuts
- temporal difference
- convex functions
- optimization problems
- optimal control
- supervised learning
- gradient field
- convex programming
- image sequences