A reinforcement learning approach to the shepherding task using SARSA.
Clark Kendrick GoBryan LaoJunichiro YoshimotoKazushi IkedaPublished in: IJCNN (2016)
Keyphrases
- reinforcement learning
- reinforcement learning algorithms
- function approximation
- function approximators
- temporal difference learning
- temporal difference
- model free
- state space
- mountain car
- rl algorithms
- action selection
- control problems
- markov decision processes
- eligibility traces
- learning classifier systems
- optimal policy
- dynamic programming
- learning algorithm
- learning problems
- reinforcement learning methods
- support vector
- markov decision process
- partially observable
- learning process
- transfer learning
- machine learning
- policy evaluation
- policy search
- evaluation function
- data sets