Learning when to observe: A frugal reinforcement learning framework for a high-cost world.
Colin BellingerMark CrowleyIsaac TamblynPublished in: CoRR (2023)
Keyphrases
- high cost
- reinforcement learning
- learning algorithm
- learning process
- supervised learning
- distributed learning
- neural network
- function approximation
- machine learning
- learning scheme
- learning systems
- autonomous learning
- prior knowledge
- learning capabilities
- learning agents
- actor critic
- kernel learning
- continuous state and action spaces
- model free
- inductive inference
- multi task
- learning problems
- knowledge acquisition
- state space