All machine learning algorithms that correspond to supervised and semi-supervised learning work under a common assumption: training and test data follow the same distribution. When the distribution changes, most statistical models must be reconstructed from new data that may be costly or even impossible to get for some applications. It therefore becomes necessary to develop approaches that reduce the need for obtaining new labeled samples. This is accomplished by exploiting data available in related areas and using it further in similar fields.
This has given rise to a new family of machine learning algorithms called transfer learning: a learning setting inspired by the capability of a human being to extrapolate knowledge across tasks to learn more efficiently. This book provides an overview of the state-of-the-art theoretical results in a specific – and arguably the most popular – subfield of transfer learning called domain adaptation.
1. State of the Art of Statistical Learning Theory.
2. Domain Adaptation Problem.
3. Seminal Divergence-based Generalization Bounds.
4. Impossibility Theorems for Domain Adaptation.
5. Generalization Bounds with Integral Probability Metrics.
6. PAC–Bayesian Theory for Domain Adaptation.
7. Domain Adaptation Theory Based on Algorithmic Properties.
8. Iterative Domain Adaptation Methods.
Ievgen Redko is Assistant Professor at Jean Monnet University in Saint-Étienne, France. His areas of interest are transfer learning, unsupervised learning and their applications.
Amaury Habrard is Professor of Computer Science at Jean Monnet University. He is head of the Data Intelligence team, and is interested in learning theory, representation and transfer learning.
Emilie Morvant is Assistant Professor in Machine Learning at Jean Monnet University. She has a special interest in theoretical machine learning, transfer learning and representation learning.
Marc Sebban is Professor of Computer Science at the Jean Monnet University and Deputy Director of the Hubert Curien CNRS laboratory. His research interests include learning from imbalanced data, metric learning and domain adaptation.
Younès Bennani is Professor of Machine Learning and Data Science at Paris 13 University. His areas of expertise are unsupervised learning, transfer and collaborative learning.
Table of Contents
PDF File 55 Kb