Transfer learning by mapping and revising boosted relational dependency networks

Detalhes bibliográficos
Ano de defesa: 2019
Autor(a) principal: Santos, Rodrigo Azevedo
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: eng
Instituição de defesa: Universidade Federal do Rio de Janeiro
Brasil
Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia
Programa de Pós-Graduação em Engenharia de Sistemas e Computação
UFRJ
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: http://hdl.handle.net/11422/13544
Resumo: Statistical machine learning algorithms usually assume that there is considerablysize data to train the models. However, traditional approaches fail to address domains where data is difficult or expensive to obtain. Transfer learning has emerged to address this problem of data scarcity by relying on a model learned in a source domain where data is easy to obtain to be a starting point for the target domain. On the other hand, real-world data is composed of objects and their relations usually disposed of in a noisy environment. Finding patterns through such uncertain relational data has been the focus of the Statistical Relational Learning area. To address these issues, scarce, relational, and uncertain data, in this work we propose TreeBoostler, an algorithm that transfers Boosted Relational Dependency Networks learned in a source domain to the target domain. TreeBoostler first finds a mapping between pairs of predicates to accommodate the trees in the target vocabulary. Then, it employs two novel theory revision operators devised to change relational regression trees to handle incorrectness and improve the performance of the mapped trees. TreeBoostler has successfully transferred knowledge among several distinct domains. It performs comparably or better than learning from scratch methods in terms of accuracy and outperforms an existing transfer learning approach in terms of accuracy and runtime.