A Convex Optimization Approach to Dynamic Programming in Continuous State and Action Spaces.
Insoon YangPublished in: J. Optim. Theory Appl. (2020)
Keyphrases
- convex optimization
- dynamic programming
- continuous state and action spaces
- reinforcement learning
- markov decision problems
- continuous state
- action selection
- low rank
- convex relaxation
- state space
- total variation
- function approximation
- optimal policy
- markov decision processes
- linear programming
- multistage
- convex optimization problems
- function approximators
- learning algorithm
- policy search
- partially observable markov decision processes
- denoising
- infinite horizon
- learning problems
- linear program