Wasserstein distributionally robust regret-optimal control over infinite-horizon.
Taylan KarginJoudi HajarVikrant MalikBabak HassibiPublished in: L4DC (2024)
Keyphrases
- optimal control
- infinite horizon
- dynamic programming
- finite horizon
- control strategy
- reinforcement learning
- robust optimization
- production planning
- average cost
- stochastic demand
- optimal control problems
- partially observable
- single item
- markov decision process
- long run
- optimal policy
- lower bound
- total reward
- lead time
- state space
- decision making
- real time