Parameter tuning of a choice-function based hyperheuristic using Particle Swarm Optimization

Crawford, Broderick; Soto, Ricardo; Monfroy, Éric; Palma, Wenceslao; Castro, Carlos; Paredes, Fernando

doi:10.1016/j.eswa.2012.09.013

Cited by 79 publications

(31 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although the tuning problem is being studied in other areas frequently [5,7,6,8], there are not many successful reports of tuning techniques in SBST. Arcuri et al tried to find a tuned setting for EvoSuite that works better than its default for a collection of classes (SF100).…”

Section: Introductionmentioning

confidence: 99%

Revisiting Hyper-Parameter Tuning for Search-Based Test Data Generation

Zamani

Hemmati

2019

Search-Based Software Engineering

View full text Add to dashboard Cite

Search-based software testing (SBST) has been studied a lot in the literature, lately. Since, in theory, the performance of metaheuristic search methods are highly dependent on their parameters, there is a need to study SBST tuning. In this study, we partially replicate a previous paper on SBST tool tuning and revisit some of the claims of that paper. In particular, unlike the previous work, our results show that the tuning impact is very limited to only a small portion of the classes in a project. We also argue the choice of evaluation metric in the previous paper and show that even for the impacted classes by tuning, the practical difference between the best and an average configuration is minor. Finally, we will exhaustively explore the search space of hyperparameters and show that half of the studied configurations perform the same or better than the baseline paper's default configuration.

show abstract

Section: Introductionmentioning

confidence: 99%

Revisiting Hyper-Parameter Tuning for Search-Based Test Data Generation

Zamani

Hemmati

2019

Search-Based Software Engineering

View full text Add to dashboard Cite

show abstract

“…Como trabajo futuro se espera poder integrar mecanismos de búsqueda autónoma en el proceso de resolución, ya que en otros trabajos han presentados excelentes resultados [9]. …”

Section: Conclusionesunclassified

Max-Min Ant System to solve the software project scheduling problem

Crawford

Soto

Johnson

et al. 2014

2014 9th Iberian Conference on Information Systems and Technologies (CISTI)

Self Cite

View full text Add to dashboard Cite

Resumen-El problema de programación de proyectos de software, es un caso particular dentro de los problemas de programación de proyectos. Este problema consiste en asignar trabajadores, con distintas habilidades, a una serie de tareas propias del proyecto de software, de manera tal que el proyecto sea completado en el menor tiempo posible y al menor costo. Además, dicha asignación debe cumplir con las restricciones de habilidad requerida por cada tareas y las restricciones de precedencia entre tarea. Se presenta la solución a este problema utilizando el algoritmo de optimización de Sistema de Hormigas Max-Min y el marco de trabajo de Hipercubo. Se proponen dos heurísticas capaces de guiar el algoritmo para busca de mejores soluciones. Se presentan los resultados experimentales y se comparan con otras técnicas de resolución, demostrando una mejora mediante la implementación de nuestra propuesta. Palabras Claves; Programación de proyectos de software, Sistemas de Hormigas Max-Min, Metaheuísticas, Hipercubo.Abstract -The Software Project Scheduling Problem is a special case of project scheduling problems. This problem consists in assigning workers with different skills, to a series of tasks in a software project so that the project is completed in the shortest time and the lowest cost possible. In addition, this allocation must meet the constraints of skill required for each task and the precedence constraints between tasks. The solution to this problem is presented using the Max -Min Ant System algorithm and the Hypercube framework. Two heuristics capable of guiding the algorithm to find better solutions are proposed. Experimental results are presented and compared with other techniques of resolution, showing an improvement by implementing our proposal.

show abstract

“…There are also many selection mechanisms which incorporate learning mechanisms. Choice Function (CF) [20,22,51] is one of the learning heuristic selection mechanisms which has been shown to perform well. This method is a score based approach in which heuristics are adaptively ranked based on a composite score.…”

Section: Heuristic Selection and Move Acceptance Methodologiesmentioning

confidence: 99%

A tensor-based selection hyper-heuristic for cross-domain heuristic search

Asta

Özcan

2015

Information Sciences

View full text Add to dashboard Cite

Hyper-heuristics have emerged as automated high level search methodologies that manage a set of low level heuristics for solving computationally hard problems. A generic selection hyper-heuristic combines heuristic selection and move acceptance methods under an iterative single point-based search framework. At each step, the solution in hand is modified after applying a selected heuristic and a decision is made whether the new solution is accepted or not. In this study, we represent the trail of a hyper-heuristic as a third order tensor. Factorization of such a tensor reveals the latent relationships between the low level heuristics and the hyper-heuristic itself. The proposed learning approach partitions the set of low level heuristics into two equal subsets where heuristics in each subset are associated with a separate move acceptance method. Then a multi-stage hyper-heuristic is formed and while solving a given problem instance, heuristics are allowed to operate only in conjunction with the associated acceptance method at each stage. To the best of our knowledge, this is the first time tensor analysis of the space of heuristics is used as a data science approach to improve the performance of a hyper-heuristic in the prescribed manner. The empirical results across six different problem domains from a benchmark indeed indicate the success of the proposed approach.

show abstract

Parameter tuning of a choice-function based hyperheuristic using Particle Swarm Optimization

Cited by 79 publications

References 6 publications

Revisiting Hyper-Parameter Tuning for Search-Based Test Data Generation

Revisiting Hyper-Parameter Tuning for Search-Based Test Data Generation

Max-Min Ant System to solve the software project scheduling problem

A tensor-based selection hyper-heuristic for cross-domain heuristic search

Contact Info

Product

Resources

About