Manifold Regularized Reinforcement Learning.
Hongliang LiDerong LiuDing WangPublished in: IEEE Trans. Neural Networks Learn. Syst. (2018)
Keyphrases
- reinforcement learning
- function approximation
- optimal policy
- state space
- reinforcement learning algorithms
- model free
- robotic control
- least squares
- temporal difference
- low dimensional
- markov decision processes
- temporal difference learning
- euclidean space
- high dimensional
- function approximators
- lower dimensional
- manifold learning
- image set
- dynamic programming
- geometric structure
- optimal control
- manifold structure
- graph embedding
- learning problems
- risk minimization
- riemannian metric
- policy search