An investigation of transfer learning for deep architectures in group activity recognition.
Karl CasserfeltRadu-Casian MihailescuPublished in: PerCom Workshops (2019)
Keyphrases
- transfer learning
- deep architectures
- group activities
- activity recognition
- learning tasks
- machine learning
- unsupervised learning
- labeled data
- cross domain
- reinforcement learning
- object recognition
- semi supervised learning
- active learning
- structure learning
- feature extraction
- pattern recognition
- text classification
- text categorization
- collaborative filtering
- action recognition
- model selection
- human activities
- transfer knowledge
- deep learning
- graphical models
- restricted boltzmann machine
- learning algorithm
- data sets
- deep belief networks
- semi supervised
- mental states
- neural nets
- e learning
- dimensionality reduction
- machine learning algorithms