Towards objective measures of algorithm performance across instance space

Smith‐Miles, Kate; Baatar, Davaatseren; Wreford, Brendan; Lewis, Rhyd

doi:10.1016/j.cor.2013.11.015

Cited by 165 publications

(141 citation statements)

References 41 publications

Supporting

Mentioning

140

Contrasting

Order By: Relevance

“…Analyzing statistical feature of classical combinatorial optimization problems and their relation to problem difficulty has gained an increasing attention in recent years [52]. Classical algorithms for the TSP and their success depending on features of the given input have been studied in [34,41,51] and similar analysis can be carried out for the knapsack problem.…”

Section: Multi-component Problemsmentioning

confidence: 99%

Evolutionary Computation for Real-World Problems

Bonyadi

Michalewicz

2015

Studies in Computational Intelligence

View full text Add to dashboard Cite

In this paper we discuss three topics that are present in the area of realworld optimization, but are often neglected in academic research in evolutionary computation community. First, problems that are a combination of several interacting sub-problems (so-called multi-component problems) are common in many real-world applications and they deserve better attention of research community. Second, research on optimisation algorithms that focus the search on the edges of feasible regions of the search space is important as high quality solutions usually are the boundary points between feasible and infeasible parts of the search space in many real-world problems. Third, finding bottlenecks and best possible investment in real-world processes are important topics that are also of interest in real-world optimization. In this chapter we discuss application opportunities for evolutionary computation methods in these three areas.

show abstract

Section: Multi-component Problemsmentioning

confidence: 99%

Evolutionary Computation for Real-World Problems

Bonyadi

Michalewicz

2015

Studies in Computational Intelligence

View full text Add to dashboard Cite

show abstract

“…There are four basic elements in the framework: (i) the problem space P represents the set of problem instances; (ii) the feature space F includes instance characteristics; (iii) the algorithm space A is the portfolio of available algorithms; and (iv) the performance space Y is the mapping of each algorithm to the performance metrics. Accordingly, the ASP can be stated as follows [87]: given a problem instance x 2 P with feature vector f .x/ 2 F , the ASP searches the selection mapping S.f .x// into algorithm space A such that the selected algorithm˛2 A maximizes the performance mapping y.˛; x/ 2 Y . Thus, for instance, [88] develops a methodology to predict the performance of metaheuristics and acquire insights into the relation between search space characteristics of an instance and algorithm performance.…”

Section: Global Hybridizationsmentioning

confidence: 99%

“…Several network architectures are assessed. In [87], the authors construct a methodology to compare the strengths and weaknesses of a set of optimization algorithms. First, the instance space is generated.…”

Section: Global Hybridizationsmentioning

confidence: 99%

Learnheuristics: hybridizing metaheuristics with machine learning for optimization with dynamic inputs

et al. 2017

View full text Add to dashboard Cite

This paper reviews the existing literature on the combination of metaheuristics with machine learning methods and then introduces the concept of learnheuristics, a novel type of hybrid algorithms. Learnheuristics can be used to solve combinatorial optimization problems with dynamic inputs (COPDIs). In these COPDIs, the problem inputs (elements either located in the objective function or in the constraints set) are not fixed in advance as usual. On the contrary, they might vary in a predictable (non-random) way as the solution is partially built according to some heuristic-based iterative process. For instance, a consumer's willingness to spend on a specific product might change as the availability of this product decreases and its price rises. Thus, these inputs might take different values depending on the current solution configuration. These variations in the inputs might require from a coordination between the learning mechanism and the metaheuristic algorithm: at each iteration, the learning method updates the inputs model used by the metaheuristic.

show abstract

“…Since Hooker's article, little has progressed in solving the difficulties that he focused on. However, in recent years, Professor SmithMiles (Smith-Miles et al, 2014;Smith-Miles and Bowly, 2015) has been developing very promising methods and techniques for data generation and analysis of the results of computational experiments.…”

mentioning

confidence: 99%

Research into the area of supply chain

Corominas

2017

View full text Add to dashboard Cite

show abstract

Towards objective measures of algorithm performance across instance space

Cited by 165 publications

References 41 publications

Evolutionary Computation for Real-World Problems

Evolutionary Computation for Real-World Problems

Learnheuristics: hybridizing metaheuristics with machine learning for optimization with dynamic inputs

Research into the area of supply chain

Contact Info

Product

Resources

About