Safety through Permissibility: Shield Construction for Fast and Safe Reinforcement Learning.
Alexander PolitowiczSahisnu MazumderBing LiuPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- machine learning
- reinforcement learning algorithms
- markov decision processes
- state space
- construction process
- dynamic programming
- markov decision process
- learning algorithm
- temporal difference
- optimal policy
- data sets
- transfer learning
- multi agent
- real time
- learning process
- optimal control
- model free
- action selection
- objective function
- case study
- robot control
- neural network
- multi agent reinforcement learning
- coal mining