Provably Safe Model-Based Meta Reinforcement Learning: An Abstraction-Based Approach.
Xiaowu SunWael FatnassiUlices Santa CruzYasser ShoukryPublished in: CDC (2021)
Keyphrases
- reinforcement learning
- model free
- state abstraction
- state space
- function approximation
- reinforcement learning algorithms
- temporal abstractions
- high level
- learning process
- machine learning
- multi agent reinforcement learning
- temporal difference
- action selection
- data driven
- optimal policy
- worst case
- multi agent
- information systems
- learning algorithm
- learning problems
- markov decision processes
- data sets
- decision making
- partially observable
- markov decision process
- data abstraction
- neural network