Model-Induced Generalization Error Bound for Information-Theoretic Representation Learning in Source-Data-Free Unsupervised Domain Adaptation.
Baoyao YangHao-Wei YehTatsuya HaradaPong C. YuenPublished in: IEEE Trans. Image Process. (2022)
Keyphrases
- information theoretic
- prior knowledge
- data sets
- probabilistic model
- training data
- learning algorithm
- active learning
- theoretic framework
- generalization error
- data points
- mutual information
- probability distribution
- supervised learning
- statistical methods
- learning machines
- learning process
- worst case
- similarity measure
- decision trees
- information theory
- image processing
- target function
- computer vision
- information theoretic measures