Use este identificador para citar ou linkar para este item:
http://repositorio.ufla.br/jspui/handle/1/40789
Título: | Optimal decision trees for the algorithm selection problem: integer programming based approaches |
Palavras-chave: | Algorithm selection problem Feature-based parameter tuning Decision trees Data mining Integer programming Variable neighborhood search COIN-OR branch and cut Algoritmo de seleção Árvores de decisão Problema de Programação Inteira Mineração de dados |
Data do documento: | Set-2019 |
Editor: | International Federation of Operational Research Societies (IFORS) |
Citação: | VILAS BOAS, M. G. et al. Optimal decision trees for the algorithm selection problem: integer programming based approaches. International Transactions in Operational Research, [S.I.], Sept. 2019. DOI: 10.1111/itor.12724 |
Resumo: | Even though it is well known that for most relevant computational problems, different algorithms may perform better on different classes of problem instances, most researchers still focus on determining a single best algorithmic configuration based on aggregate results such as the average. In this paper, we propose integer programming‐based approaches to build decision trees for the algorithm selection problem. These techniques allow the automation of three crucial decisions: (urn:x-wiley:09696016:media:itor12724:itor12724-math-0001) discerning the most important problem features to determine problem classes, (urn:x-wiley:09696016:media:itor12724:itor12724-math-0002) grouping the problems into classes, and (urn:x-wiley:09696016:media:itor12724:itor12724-math-0003) selecting the best algorithm configuration for each class. To evaluate this new approach, extensive computational experiments were executed using the linear programming algorithms implemented in the COIN‐OR branch‐and‐cut solver across a comprehensive set of instances, including all MIPLIB benchmark instances. The results exceeded our expectations. While selecting the single best parameter setting across all instances decreased the total running time by 22%, our approach decreased the total running time by 40% on average across 10‐fold cross‐validation experiments. These results indicate that our method generalizes quite well and does not overfit. |
URI: | https://onlinelibrary.wiley.com/doi/10.1111/itor.12724 http://repositorio.ufla.br/jspui/handle/1/40789 |
Aparece nas coleções: | DCC - Artigos publicados em periódicos |
Arquivos associados a este item:
Não existem arquivos associados a este item.
Os itens no repositório estão protegidos por copyright, com todos os direitos reservados, salvo quando é indicado o contrário.
Ferramentas do administrador