GreedEx and OptimEx: two tools to experiment with optimization algorithms

Autores: Jesús Angel Velázquez Iturbide
Localización: The International journal of engineering education, ISSN-e 0949-149X, Vol. 32, no. 3 (Parte A), 2016, págs. 1097-1106
Idioma: inglés
Texto completo no disponible (Saber más ...)

Resumen
- Experimentation is an important part of the education of computer engineers. In particular, it is a common educationalpractice to check algorithms correctness or efficiency. However, experimentation has seldom been used to check theoptimality of (optimization) algorithms. The article presents two tools aimed at experimenting with this property. Theyhave some common features, but differ in their degree of generality and scaffolding. GreedEx is a tool for novices, beingaimed at the active learning of the foundations of greedy algorithms. It currently supports six optimization problems.OptimEx is a more advanced, general experimentation tool that can be used with any kind of optimization algorithms. Ifboth systems are used in an algorithm course, they should be used at different stages. The paper presents two contributions.Firstly, we present the novel system OptimEx. Secondly, we give recommendations of use for both tools, based on theauthor’s experience using and evaluating them. Of particular interest is a list of incorrect outcomes that may be producedby OptimEx, which are symptoms of students’ misconceptions, as well as how to fix them.