Exploration Without Maps via Zero-Shot Out-of-Distribution Deep Reinforcement Learning.
Shathushan SivashangaranApoorva KhairnarAzim EskandarianPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- active exploration
- exploration strategy
- probability distribution
- exploration exploitation
- supervised learning
- action selection
- function approximation
- learning algorithm
- exploration exploitation tradeoff
- model based reinforcement learning
- autonomous learning
- markov decision processes
- uniformly distributed
- power law
- reinforcement learning algorithms
- temporal difference learning
- random variables
- robotic control
- multi agent
- data sets
- probability density function
- spatial distribution
- policy search
- image features
- state space
- training set