Safety-Aware Task Composition for Discrete and Continuous Reinforcement Learning.
Kevin LeahyMakai MannZachary SerlinPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- continuous domains
- continuous state spaces
- continuous state and action spaces
- continuous action
- discrete space
- continuous state
- state space
- continuous variables
- policy search
- continuous space
- function approximation
- action space
- discrete data
- continuous data
- discrete variables
- control problems
- continuous functions
- model free
- markov decision processes
- reinforcement learning algorithms
- real time
- learning algorithm
- robotic control
- learning process
- discrete geometry
- multi agent
- neural network
- discrete valued
- discrete version
- safety analysis
- sequence prediction
- multiscale
- data streams
- transfer learning
- piecewise linear
- action selection
- temporal difference