Reward-Weighted GMM and Its Application to Action-Selection in Robotized Shoe Dressing.
Adrià ColoméSergi FoixGuillem AlenyàCarme TorrasPublished in: ROBOT (2) (2017)
Keyphrases
- action selection
- reinforcement learning
- total reward
- gaussian mixture model
- basal ganglia
- robot soccer
- decision making
- human robot
- temporal difference
- function approximation
- machine learning
- gaussian mixture
- learning algorithm
- mixture model
- em algorithm
- reinforcement learning algorithms
- speaker identification
- state space
- feature vectors
- gaussian mixture modeling
- action selection mechanism
- markov decision processes
- action space