Self-Rule to Adapt: Learning Generalized Features from Sparsely-Labeled Data Using Unsupervised Domain Adaptation for Colorectal Cancer Tissue Phenotyping.
Christian AbbetLinda StuderAndreas FischerHeather DawsonInti ZlobecBehzad BozorgtabarJean-Philippe ThiranPublished in: MIDL (2021)
Keyphrases
- labeled data
- supervised learning
- active learning
- prior knowledge
- learning algorithm
- unlabeled data
- semi supervised learning
- feature engineering
- domain adaptation
- labeled and unlabeled data
- semi supervised
- supervised learning algorithms
- training data
- training examples
- sample selection bias
- class labels
- co training
- text classification
- semi supervised classification
- labeled examples
- background knowledge
- partially labeled data
- data sets
- learning models
- learning problems
- text categorization
- feature set
- learning tasks
- positive examples
- unsupervised learning
- data points
- learning process
- feature space
- reinforcement learning
- labeling effort
- feature extraction