Empowering parameter-efficient transfer learning by recognizing the kernel structure in self-attention.
Yifan ChenDevamanyu HazarikaMahdi NamazifarYang LiuDi JinDilek Hakkani-TurPublished in: NAACL-HLT (Findings) (2022)
Keyphrases
- transfer learning
- knowledge transfer
- learning tasks
- reinforcement learning
- labeled data
- semi supervised learning
- transfer knowledge
- machine learning
- text classification
- manifold alignment
- target domain
- multi task
- text categorization
- active learning
- feature space
- neural network
- semi supervised
- information extraction
- knowledge representation
- probabilistic model
- cross domain
- domain adaptation
- multi task learning
- data sets