Leticia Hernando

An evaluation of methods for estimating the number of local optima in combinatorial optimization problems

The solution of many combinatorial optimization problems is carried out by metaheuristics, which generally make use of local search algorithms. These algorithms use some kind of neighborhood structure over the search space. The performance of the algorithms strongly depends on the properties that the neighborhood imposes on the search space. One of these properties is the number of local optima. Given an instance of a combinatorial optimization problem and a neighborhood, the estimation of the number of local optima can help not only to measure the complexity of the instance, but also to choose the most convenient neighborhood to solve it. In this paper we review and evaluate several methods to estimate the number of local optima in combinatorial optimization problems. The methods reviewed not only come from the combinatorial optimization literature, but also from the statistical literature. A thorough evaluation in synthetic as well as real problems is given. We conclude by providing recommendations of methods for several scenarios.

A Tunable Generator of Instances of Permutation-Based Combinatorial Optimization Problems

In this paper, we propose a tunable generator of instances of permutation-based combinatorial optimization problems. Our approach is based on a probabilistic model for permutations, called the generalized Mallows model. The generator depends on a set of parameters that permits the control of the properties of the output instances. Specifically, in order to create an instance, we solve a linear programming problem in the parameters, where the restrictions allow the instance to have a fixed number of local optima and the linear function encompasses qualitative characteristics of the instance. We exemplify the use of the generator by giving three distinct linear functions that produce three landscapes with different qualitative properties. After that, our generator is tested in two different ways. First, we test the flexibility of the model by producing instances similar to benchmark instances. Second, we account for the capacity of the generator to create different types of instances according to the difficulty for population-based algorithms. We study the influence of the input parameters in the behaviors of these algorithms, giving an example of a property that can be used to analyze their performance.

Generating Customized Landscapes in Permutation-Based Combinatorial Optimization Problems

Designing customized optimization problem instances is a key issue in optimization. They can be used to tune and evaluate new algorithms, to compare several optimization algorithms, or to evaluate techniques that estimate the number of local optima of an instance. Given this relevance, several methods have been proposed to design customized optimization problems in the field of evolutionary computation for continuous as well as binary domains. However, these proposals have not been extended to permutation spaces. In this paper we provide a method to generate customized landscapes in permutation-based combinatorial optimization problems. Based on a probabilistic model for permutations, called the Mallows model, we generate instances with specific characteristics regarding the number of local optima or the sizes of the attraction basins.

A study on the complexity of TSP instances under the 2-exchange neighbor system

This work is related to the search of complexity measures for instances of combinatorial optimization problems. Particularly, we have carried out a study about the complexity of random instances of the Traveling Salesman Problem under the 2-exchange neighbor system. We have proposed two descriptors of complexity: the proportion of the size of the basin of attraction of the global optimum over the size of the search space and the proportion of the number of different local optima over the size of the search space. We have analyzed the evolution of these descriptors as the size of the problem grows. After that, and using our complexity measures, we find a phase transition phenomenon in the complexity of the instances.

Understanding Instance Complexity in the Linear Ordering Problem

The Linear Ordering Problem is a combinatorial optimization problem which has been frequently addressed in the literature due to its numerous applications in diverse fields. In spite of its popularity, little is known about its complexity. In this paper we analyze the linear ordering problem trying to identify features or characteristics of the instances that can provide useful insights into the difficulty of solving them. Particularly, we introduce two different metrics, insert ratio and ubiquity ratio, that measure the difficulty of solving the LOP with local search type algorithms with the insert neighborhood system. Conducted experiments demonstrate that the proposed metrics clearly correlate with the complexity of solving the LOP with a multistart local search algorithm.

Local Optima Networks of the Permutation Flowshop Scheduling Problem: Makespan vs. total flow time

Local Optima Networks were proposed to understand the structure of combinatorial landscapes at a coarse-grained level. We consider a compressed variant of such networks with features that are meaningful for the study of search difficulty in the context of local search. In particular, we investigate different landscapes of the Permutation Flowshop Scheduling Problem. The insert and 2-exchange neighbourhoods are considered, and two different objective functions are taken into account: the makespan and the total flow time. The aim is to analyse the network features in order to find differences between the landscape structures, giving insights about which features impact algorithm performance. We evaluate the correlation between landscape properties and the performance of an Iterated Local Search algorithm. Visualisation of the network structure is also given, where evident differences between the makespan and total flow time are observed.

Estimating attraction basin sizes of combinatorial optimization problems

Given a particular instance of a combinatorial optimization problem, the knowledge about the attraction basin sizes can help to analyze the difficulty encountered by local search algorithms while solving it. As calculating these sizes exhaustively is computationally intractable, we focus on methods for their estimation. The accuracy of some of these estimation methods depends on the way in which the sample of solutions of the search space is chosen. In this paper, we propose a novel sampling method, which incorporates the knowledge obtained by the already explored solutions into the sampling strategy. So, in contrast to those that already exist, our method can adapt its behavior to the characteristics of the particular attraction basin. We apply our proposal to a number of instances of three famous problems: the quadratic assignment problem, the linear ordering problem and the permutation flow shop scheduling problem. We consider permutation sizes

Estimating Attraction Basin Sizes

n = 10