Action redundancy in reinforcement learning.
Nir BaramGuy TennenholtzShie MannorPublished in: UAI (2021)
Keyphrases
- reinforcement learning
- action selection
- action space
- partially observable domains
- reward shaping
- function approximation
- state space
- reinforcement learning algorithms
- machine learning
- state action
- multi agent
- learning algorithm
- model free
- partially observable
- transition model
- optimal control
- markov decision processes
- learning process
- sensory inputs
- agent learns
- function approximators
- reasoning about actions