On Uninformative Optimal Policies in Adaptive LQR with Unknown B-Matrix.
Ingvar ZiemannHenrik SandbergPublished in: CoRR (2020)
Keyphrases
- optimal policy
- markov decision processes
- decision problems
- dynamic programming
- finite horizon
- reinforcement learning
- state space
- long run
- infinite horizon
- finite state
- dynamic programming algorithms
- average reward
- serial inventory systems
- optimal control
- state dependent
- multistage
- average cost
- initial state
- markov decision process
- policy iteration
- expected cost
- linear programming
- semi markov decision processes
- average reward reinforcement learning
- partially observable markov decision processes
- computational complexity
- data mining