Infinite Horizon Average Cost Dynamic Programming Subject to Total Variation Distance Ambiguity.
Ioannis TzortzisCharalambos D. CharalambousThemistoklis CharalambousPublished in: SIAM J. Control. Optim. (2019)
Keyphrases
- infinite horizon
- total variation
- average cost
- dynamic programming
- image restoration
- denoising
- image denoising
- finite horizon
- optimal policy
- optimal control
- long run
- markov decision processes
- single item
- state space
- production planning
- multistage
- markov decision process
- partially observable
- dec pomdps
- image processing
- inventory models
- initial state
- holding cost
- finite state
- lead time
- inventory control
- lost sales
- inventory level
- policy iteration
- fixed cost
- stationary policies
- knapsack problem
- search algorithm
- setup cost
- natural images
- markov decision problems