Suggestion of probabilistic reward-independent knowledge for dynamic environment in reinforcement learning.
Nodoka ShibuyaYoshiki MiyazakiKentarou KurashigePublished in: MHS (2011)
Keyphrases
- dynamic environments
- reinforcement learning
- mobile robot
- path planning
- reinforcement learning algorithms
- domain knowledge
- autonomous agents
- reinforcement learning agents
- bayesian networks
- knowledge base
- knowledge acquisition
- learning agent
- probabilistic knowledge
- expert systems
- multi agent
- real environment
- knowledge representation
- adaptive control
- machine learning
- learning process
- function approximation
- collision avoidance
- path planner
- real time