Parametric UMAP: learning embeddings with deep neural networks for representation and semi-supervised learning.
Tim SainburgLeland McInnesTimothy Q. GentnerPublished in: CoRR (2020)
Keyphrases
- semi supervised learning
- supervised learning
- neural network
- active learning
- unlabeled data
- learning problems
- semi supervised
- unsupervised learning
- labeled and unlabeled data
- learning models
- co training
- learning algorithm
- labeled data
- multi view learning
- learning process
- graph transduction
- machine learning
- reinforcement learning
- supervised learning algorithms
- label propagation
- regularization framework
- semi supervised classification
- domain adaptation
- learning tasks
- data sets
- manifold regularization
- training set
- training data
- boosting framework
- covariate shift
- learning experience
- face recognition
- class labels
- decision trees
- prior knowledge
- feature extraction