Learning a bidirectional mapping between human whole-body motion and natural language using deep recurrent neural networks.
Matthias PlappertChristian ManderyTamim AsfourPublished in: CoRR (2017)
Keyphrases
- recurrent neural networks
- recurrent networks
- learning process
- natural language
- cascade correlation
- neural network
- reinforcement learning
- learning tasks
- real time
- learning algorithm
- motion estimation
- feedforward neural networks
- long short term memory
- expressed in natural language
- language acquisition
- humanoid robot
- knowledge representation
- long term
- prior knowledge
- artificial neural networks
- machine learning