Optimal decision trees for the algorithm selection problem: integer programming based approaches

doi:10.1111/itor.12724

Optimal decision trees for the algorithm selection problem: integer programming based approaches

2019-09-23

13

OA

AI

V

Vilas Boas, Matheus Guedes *

S

Santos, Haroldo Gambini

L

Luiz Henrique de Campos Merschmann

G

Greet Vanden Berghe

DOI：10.1111/itor.12724

原文链接

查看原文

摘要

En 中文

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: (a) discerning the most important problem features to determine problem classes, (b) grouping the problems into classes, and (c) 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.

Keyword:

algorithm selection problem

feature-based parameter tuning

decision trees

data mining

integer programming

variable neighborhood search

COIN-OR branch and cut

Optimal decision trees for the algorithm selection problem: integer programming based approaches

摘要

AI总结

期刊

机构