CAMEO: Curiosity Augmented Metropolis for Exploratory Optimal Policies.
Mohamed Alami ChehbouneFernando LlorenteRim KaddahLuca MartinoJesse ReadPublished in: CoRR (2022)
Keyphrases
- optimal policy
- markov decision processes
- dynamic programming
- decision problems
- infinite horizon
- state space
- simulated annealing
- finite horizon
- finite state
- reinforcement learning
- state dependent
- multistage
- long run
- bayesian reinforcement learning
- markov decision process
- average reward
- sufficient conditions
- policy iteration
- serial inventory systems
- initial state
- control policies
- dynamic programming algorithms
- average reward reinforcement learning
- lost sales
- average cost
- finite number