Approximate Model-Based Shielding for Safe Reinforcement Learning.
Alexander W. GoodallFrancesco BelardinelliPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- model free
- function approximation
- policy evaluation
- reinforcement learning algorithms
- multi agent
- robotic control
- markov decision processes
- data driven
- state space
- learning process
- optimal policy
- temporal difference
- reward function
- hidden markov models
- supervised learning
- data mining
- learning problems
- artificial neural networks
- case study
- information systems
- continuous state
- social networks
- policy search
- artificial intelligence