Augmented Lagrangian Method for Instantaneously Constrained Reinforcement Learning Problems.
Jingqi LiDavid Fridovich-KeilSomayeh SojoudiClaire J. TomlinPublished in: CDC (2021)
Keyphrases
- reinforcement learning problems
- augmented lagrangian method
- reinforcement learning algorithms
- augmented lagrangian
- total variation
- reinforcement learning methods
- reinforcement learning
- convex optimization
- total variation regularization
- function approximators
- function approximation
- policy iteration
- constrained minimization
- temporal difference
- markov decision problems
- markov decision processes
- lagrange multipliers
- image processing
- cost function
- machine learning
- denoising