Learning from Label Proportions via an Iterative Weighting Scheme and Discriminant Analysis

Áreas de investigación:

• Methodology - Machine learning

Año:

2016

Tipo de publicación:

Artículo en conferencia

Autores:

• Pérez-Ortiz, María
• Gutiérrez, Pedro Antonio
• Carbonero-Ruz, Mariano
• Hervás-Martínez, César

Volumen:

9868

Título del libro:

Proceedings of the 17th Conference of the Spanish Association for Artificial Intelligence (CAEPIA 2016)

Serie:

Lecture Notes on Computer Science (LNCS)

Páginas:

79-88

Organización:

Salamanca, Spain

Mes:

Septiembre, 2016

ISBN:

978-3-319-44635-6

ISSN:

0302-9743

BibTex:

@conference{Maria2016CAEPIA, author = "Mar{\'i}a P{\'e}rez-Ortiz and Pedro Antonio Guti{\'e}rrez and Mariano Carbonero-Ruz and C{\'e}sar Herv{\'a}s-Mart{\'i}nez", abstract = "Learning from label proportions is the term used for the learning paradigm where the training data is provided in groups (or {\&}#039;bags{\&}#039;), and only the label proportion for each bag is known. The objective is to learn a model to predict the class labels of individual instances. This paradigm presents very different applications, specially concerning anonymous data. Two different iterative strategies are proposed to deal with this type of problems, both based on optimising the class membership of the instances using the estimated pattern distribution per bag and the label proportions. Discriminant analysis is reformulated to deal with non-crisp class memberships. A thorough set of experiments is conducted to test: (1) the performance gap between these approaches and the fully supervised setting, (2) the potential advantages of optimising class memberships by our proposals, and (3) the influence of factors such as the bag size and the number of classes of the problem in the performance.", booktitle = "Proceedings of the 17th Conference of the Spanish Association for Artificial Intelligence (CAEPIA 2016)", doi = "10.1007/978-3-319-44636-3_8", isbn = "978-3-319-44635-6", issn = "0302-9743", month = "Septiembre, 2016", organization = "Salamanca, Spain", pages = "79-88", publisher = "Springer International Publishing", series = "Lecture Notes on Computer Science (LNCS)", title = "{L}earning from {L}abel {P}roportions via an {I}terative {W}eighting {S}cheme and {D}iscriminant {A}nalysis", url = "dx.doi.org/10.1007/978-3-319-44636-3_8", volume = "9868", year = "2016", }

Abstract:

Learning from label proportions is the term used for the learning paradigm where the training data is provided in groups (or 'bags'), and only the label proportion for each bag is known. The objective is to learn a model to predict the class labels of individual instances. This paradigm presents very different applications, specially concerning anonymous data. Two different iterative strategies are proposed to deal with this type of problems, both based on optimising the class membership of the instances using the estimated pattern distribution per bag and the label proportions. Discriminant analysis is reformulated to deal with non-crisp class memberships. A thorough set of experiments is conducted to test: (1) the performance gap between these approaches and the fully supervised setting, (2) the potential advantages of optimising class memberships by our proposals, and (3) the influence of factors such as the bag size and the number of classes of the problem in the performance.