Quasi-Newton Iteration in Deterministic Policy Gradient.
Arash Bahari KordabadHossein Nejatbakhsh EsfahaniWenqi CaiSebastien GrosPublished in: CoRR (2022)
Keyphrases
- quasi newton
- policy gradient
- gradient method
- step size
- optimization methods
- optimization method
- function approximation
- newton method
- reinforcement learning
- reinforcement learning algorithms
- approximation methods
- neural network
- objective function
- genetic algorithm
- partially observable markov decision processes
- convergence rate
- optimization algorithm
- radial basis function
- learning tasks
- function approximators
- optimization problems
- principal component analysis
- dynamic programming
- multi objective
- decision trees