Leverage Unlabeled Data for Abstractive Speech Summarization with Self-supervised Learning and Back-Summarization.
Paul TardyLouis de SeynesFrançois HernandezVincent NguyenDavid JaniszekYannick EstèvePublished in: SPECOM (2020)
Keyphrases
- unlabeled data
- active learning
- learning algorithm
- supervised learning
- semi supervised learning
- semi supervised
- co training
- labeled and unlabeled data
- labeled data
- domain adaptation
- positive examples
- prior knowledge
- learning process
- supervised learning algorithms
- learning frameworks
- reinforcement learning
- sample selection bias
- labeling effort
- machine learning
- semi supervised classification
- transfer learning
- unsupervised learning
- data points
- support vector
- class labels
- negative examples
- text mining
- training data
- data mining
- small set of labeled