A Cubic-regularized Policy Newton Algorithm for Reinforcement Learning.
Mizhaan Prajit ManiyarPrashanth L. A.Akash MondalShalabh BhatnagarPublished in: AISTATS (2024)
Keyphrases
- learning algorithm
- detection algorithm
- reinforcement learning
- optimal solution
- preprocessing
- significant improvement
- dynamic programming
- cost function
- computational complexity
- objective function
- search space
- neural network
- optimization algorithm
- model free
- total least squares
- k means
- monte carlo
- multi agent
- computational cost
- simulated annealing
- image segmentation
- convergence rate