Matryoshka Policy Gradient for Entropy-Regularized RL: Convergence and Global Optimality.
François GedMaria Han VeigaPublished in: CoRR (2023)
Keyphrases
- reinforcement learning methods
- policy gradient
- global optimality
- reinforcement learning
- globally optimal
- reinforcement learning algorithms
- global optimization
- actor critic
- optimal solution
- objective function
- model free
- global minimum
- function approximation
- state space
- theoretical guarantees
- convex functions
- convergence rate
- learning algorithm
- markov decision processes
- function approximators
- particle swarm optimization
- np hard
- cost function
- lower bound
- rl algorithms
- multi agent
- image segmentation