Generic resources are what you need: Style transfer tasks without task-specific parallel training data.
Huiyuan LaiAntonio ToralMalvina NissimPublished in: EMNLP (1) (2021)
Keyphrases
- training data
- transfer learning
- previously learned
- resource allocation
- limited resources
- learning algorithm
- prior knowledge
- computational grids
- domain knowledge
- parallel processing
- domain specific
- test data
- knowledge transfer
- generalization error
- transferring knowledge
- parallel implementation
- parallel computing
- multiple tasks
- data sets
- multi task
- semi supervised learning
- labeled data
- data management
- supervised learning
- classification accuracy
- training set
- database systems
- neural network