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