Reinforcement Learning in Linear Quadratic Deep Structured Teams: Global Convergence of Policy Gradient Methods.
Vida FathiJalal ArabneydiAmir G. AghdamPublished in: CoRR (2020)
Keyphrases
- global convergence
- policy gradient methods
- linear quadratic
- optimal control
- reinforcement learning
- actor critic
- policy gradient
- natural actor critic
- convergence rate
- global optimum
- convergence speed
- optimization methods
- closed loop
- vector valued
- multi agent
- dynamic programming
- gradient method
- robot arm
- dynamical systems
- machine learning
- particle swarm
- reinforcement learning algorithms
- function approximators
- gaussian model
- state space
- function approximation
- reinforcement learning methods
- optimal policy
- optimization problems
- step size
- rl algorithms
- evolutionary algorithm
- reinforcement learning problems
- simulated annealing
- real valued
- particle swarm optimization
- optimization method
- markov decision processes
- learning algorithm
- genetic algorithm