A Unified Framework for Domain Adaptation Using Metric Learning on Manifolds.
Sridhar MahadevanBamdev MishraShalini GhoshPublished in: ECML/PKDD (2) (2018)
Keyphrases
- metric learning
- domain adaptation
- semi supervised
- manifold alignment
- manifold learning
- manifold structure
- semi supervised learning
- labeled data
- learning tasks
- transfer learning
- unlabeled data
- distance metric
- feature space
- pairwise
- multi task
- dimensionality reduction
- multiple sources
- active learning
- unsupervised learning
- laplacian eigenmaps
- co training
- supervised learning
- target domain
- cross domain
- sentiment classification
- kernel matrix
- distance function
- euclidean space
- text classification
- low dimensional
- high dimensional data
- knowledge transfer
- information retrieval
- high dimensional
- support vector
- reinforcement learning
- feature selection