Publications

2019

1. SoftFEM: The Soft Finite Element Method Peña, Jose M., LaTorre, Antonio, and Jérusalem, Antoine International Journal for Numerical Methods in Engineering 2019 [Abs] [HTML]

   While the finite element method (FEM) has now reached full maturity both in academy and industry, its use in optimisation pipelines remains either computationally intensive or cumbersome. In particular, currently used optimisation schemes leveraging FEM still require the choice of dedicated optimisation algorithms for a specific design problem, and a “black box” approach to FEM-based optimisation remains elusive. To this end, we propose here an integrated finite element-soft computing method, the Soft Finite Element Method (SoftFEM), which integrates a finite element solver within a metaheuristic search wrapper. To illustrate this general method, we focus here on solid mechanics problems. For these problems, SoftFEM is able to optimise geometry changes and mechanistic measures based on geometry constraints and material properties inputs. From the optimisation perspective, the use of a fitness function based on finite element calculation imposes a series of challenges. To bypass the limitations in search capabilities of the usual optimisation techniques (local search and gradient-based methods), we propose instead a hybrid self adaptive search technique, the Multiple Offspring Sampling (MOS), combining two metaheuristics methods: one population-based differential evolution method and a local search optimiser. The formulation coupling FEM to the optimisation wrapper is presented in detail and its flexibility is illustrated with three representative solid mechanics problems. More particularly, we propose here the MOS as the most versatile search algorithm for SoftFEM. A new method for the identification of non-fully determined parameters is also proposed.

2. InTool Explorer: An Interactive Exploratory Analysis Tool for Versatile Visualizations of Neuroscientific Data Furcila, Diana, García, Marcos, Toader, Cosmin, Morales, Juan, LaTorre, Antonio, Rodriguez, Angel, Pastor, Luis, DeFelipe, Javier, and Alonso-Nanclares, Lidia Frontiers in Neuroanatomy 2019 [Abs] [HTML]

   The bottleneck for progress in many research areas within neuroscience has shifted from the data acquisition to the data analysis stages. In the present article, we propose a method named InTool Explorer that we have developed to perform interactive exploratory data analysis, focusing on neuroanatomy as an example of its utility. This tool is freely-available software that has been designed to facilitate the study of complex neuroscience data. InTool Explorer requires no more than an internet connection and a web browser. The main contribution of this tool is to provide a user-designed canvas for data visualization and interaction, to perform specific exploratory tasks according to the user needs. Moreover, InTool Explorer permits visualization of the datasets in a very dynamic and versatile way using a linked-card approach. For this purpose, the tool allows the user to select among different predefined card types. Each card type offers an abstract data representation, a filtering tool or a set of statistical analysis methods. Additionally, InTool Explorer makes it possible linking raw images to the data. These images can be used by InTool Explorer to define new customized filtering cards. Another significant contribution of this tool is that it allows fast visualization of the data, error finding, and re-evaluation to establish new hypotheses or new lines of research. Thus, regarding its practical application in the laboratory, InTool Explorer provides a new opportunity to study and analyze neuroscience data prior to any statistical analysis being carried out.

2018

1. An Insight into Bio-Inspired and Evolutionary Algorithms for Global Optimization: Review, Analysis, and Lessons Learnt over a Decade of Competitions Molina, Daniel, LaTorre, Antonio, and Herrera, Francisco Cognitive Computation 2018 [Abs] [HTML]

   Over the recent years, continuous optimization has significantly evolved to become the mature research field it is nowadays. Through this process, evolutionary algorithms had an important role, as they are able to obtain good results with limited resources. Among them, bio-inspired algorithms, which mimic cooperative and competitive behaviors observed in animals, are a very active field, with more proposals every year. This increment in the number of optimization algorithms is apparent in the many competitions held at corresponding special sessions in the last 10 years. In these competitions, several algorithms or ideas have become points of reference, and used as starting points for more advanced algorithms in following competitions. In this paper, we have obtained, for different real-parameter competitions, their benchmarks, participants, and winners (from the competitions’ website) and we review the most relevant algorithms and techniques, presenting the trajectory they have followed over the last years and how some of these works have deeply influenced the top performing algorithms of today. The aim is to be both a useful reference for researchers new to this interesting research topic and a useful guide for current researchers in the field. We have observed that there are several algorithms, like the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), the Success-History based Adaptive Differential Evolution with Linear Population Size Reduction (L-SHADE), Mean-Variance Mapping Optimization (MVMO), and Multiple Offspring Sampling (MOS), which have obtained a strong influence over other algorithms. We have also suggested several techniques that are being widely adopted among the winning proposals, and that could be used for more competitive algorithms. Global optimization is a mature research field in continuous improvement, and the history of competitions provides useful information about the past that can help us to learn how to go forward in the future.

2. Toolkit for the Automatic Comparison of Optimizers: Comparing Large-Scale Global Optimizers Made Easy Molina, D., and LaTorre, A. In 2018 IEEE Congress on Evolutionary Computation (CEC) 2018 [Abs]

   Large-scale global optimization is a research subject that has attracted significant attention, including both theoretical and practical studies, in recent years. In this sense, some of the main conferences in the field of Evolutionary Computation have been organizing special sessions on this topic for more than a decade. Those special sessions normally propose a well defined benchmark of functions to allow a fair comparison of participating algorithms. Being able to manually tackle all this information has become a difficult task for many researchers. Which algorithm obtained the best results on a particular benchmark? How does the new method that I am developing compare to that algorithm? To answer this question, we propose Toolkit for the Automatic Comparison of Optimizers, TACO, a web application that stores all this information and makes it possible to seamlessly analyze it and generate detailed reports with the results of these analyses. The application has been designed in such a flexible way that it is extremely easy to add new benchmarks and their associated (possibly specific) analyses. Of course, these benchmarks are not limited to large-scale global optimization, but potentially any type of optimization problems. Finally, we also provide a publicly accessible instance demonstrating the features of the application with ready-to-use results from some recent special sessions.

3. SHADE with Iterative Local Search for Large-Scale Global Optimization Molina, D., LaTorre, A., and Herrera, F. In 2018 IEEE Congress on Evolutionary Computation (CEC) 2018 [Abs]

   Global optimization is a very important topic in research due to its wide applications in many real-world problems in science and engineering. Among optimization problems, dimensionality is one of the most crucial issues that increases the difficulty of the optimization process. Thus, Large-Scale Global Optimization, optimization with a great number of variables, arises as a field that is getting an increasing interest. In this paper, we propose a new hybrid algorithm especially designed to tackle this type of optimization problems. The proposal combines, in a iterative way, a modern Differential Evolution algorithm with one local search method chosen from a set of different search methods. The selection of the local search method is dynamic and takes into account the improvement obtained by each of them in the previous intensification phase, to identify the most adequate in each case for the problem. Experiments are carried out using the CEC’2013 Large-Scale Global Optimization benchmark, and the proposal is compared with other state-of-the-art algorithms, showing that the synergy among the different components of our proposal leads to better and more robust results than more complex algorithms. In particular, it improves the results of the current winner of previous Large-Scale Global Optimization competitions, Multiple Offspring Sampling, MOS, obtaining very good results, especially in the most difficult problems.

4. Toolkit for the Automatic Comparison of Optimizers (TACO): Herramienta online avanzada para comparar metaheurísticas Molina, Daniel, and LaTorre, Antonio In XIII Congreso Español en Metaheurísticas y Algoritmos Evolutivos y Bioinspirados, XVIII Conferencia de la Asociación Española para la Inteligencia Artificial 2018 [Abs] [HTML]

   El campo de las metaheurísticas es un campoen continua innovación, en donde los investigadores proponenfrecuentemente nuevas propuestas, propuestas que tienen que seradecuadamente comparadas con el estado-del-arte para probarsu efectividad. El uso de benchmarks estándar permite comparardirectamente los resultados de los algoritmos entre sí, peroincluso esas comparaciones suponen un proceso laborioso yrepetitivo que consume gran parte del proceso de investigaciónen metaheurísticas. En este trabajo proponemos Toolkit for theAutomatic Comparison of Optimizers, TACO, una herramientaonlineavanzada para comparar metaheurísticas, que además dealmacenar los resultados de distintos algoritmos competitivospara distintos benchmarks (para usarlos como algoritmos dereferencia), es capaz de realizar de forma automática unavariedad de comparaciones. De hecho, permite comparar nosólo los algoritmos previamente almacenados, sino también losresultados de nuevos algoritmos sin necesidad de registrarse, eincluso para funciones específicas (no limitado a un benchmarkconcreto). Consideramos que es una herramienta muy útil paratodo tipo de investigaciones en metaheurísticas. Como pruebade su conveniencia, ofrecemos una instancia de demostración desus características, sobre la que se han realizado las distintascomparaciones que incluimos en este trabajo.

5. SHADE con Búsqueda Local Iterativa para Optimización Global de Alta Dimensionalidad Molina, Daniel, LaTorre, Antonio, and Herrera, Francisco In XIII Congreso Español en Metaheurísticas y Algoritmos Evolutivos y Bioinspirados, XVIII Conferencia de la Asociación Española para la Inteligencia Artificial 2018 [HTML]
6. An Insight into Evolutionary Algorithms for Continuous Optimization: Learning by Competitions Molina, Daniel, LaTorre, Antonio, and Herrera, Francisco In XIII Congreso Español En Metaheurísticas y Algoritmos Evolutivos y Bioinspirados, XVIII Conferencia de La Asociación Española Para La Inteligencia Artificial 2018 [Abs] [HTML]

   During the recent years, continuous optimization has significantly evolved to become a mature research field, in which evolutionary and bio-inspired algorithms have an important role thanks to their good results with limited resources. Through the last ten years, there have been a number of competitions associated to special sessions in this area. In this study, we summarize, for different real-parameter competitions, the results, highlighting the most relevant algorithms and techniques, presenting the trajectory they have followed and how some of these works have deeply influenced the top performing algorithms of today. The aim is to be both a useful reference for researchers new to this interesting research topic and a useful guide for current researchers in the field.

7. SoftFEM: The Soft Finite Element Method Peña, J M, LaTorre, A, and Jérusalem, A In 6th European Conference on Computational Mechanics (Solids, Structures and Coupled Problems) (ECCM 6) and 7th European Conference on Computational Fluid Dynamics (ECFD 7) 2018 [HTML]
8. SoftFEM: The Soft Finite Element Method Peña, J M, LaTorre, A, and Jérusalem, A In Society of Engineering’s 55th Annual Technical Meeting SES2018 2018 [HTML]

2017

1. On the Scalability of Population Restart Mechanisms on Large-Scale Global Optimization LaTorre, A, and Peña, J M In 2017 IEEE Congress on Evolutionary Computation (CEC 2017) 2017 [Abs] [HTML]

   Population restart mechanisms are a popular method to avoid premature convergence in Evolutionary Algorithms. Many different methods have used these mechanisms in the past in different scenarios. However, most of these works tend to design an ad-hoc population restart approach for the problem under consideration. Furthermore, the effects of the alternative restart strategies and the scalability of the method are rarely analyzed in the literature. In this paper, we conduct a comparative study of 36 population restart strategies (37 if we account for the baseline of not restarting the population) on the SOCO 2011 benchmark, a testbed of 19 continuous scalable functions widely accepted in the continuous optimisation community and that allow an analysis at different problem sizes which is not possible with other existing benchmarks. The results obtained clearly show that there is a relationship between the particular strategy considered and the effectiveness of the method. Moreover, this effectiveness tends to decrease as the dimensionality (complexity) of the problem grows.

2. A Comparison of Three Large-Scale Global Optimizers on the CEC 2017 Single Objective Real Parameter Numerical Optimization Benchmark LaTorre, A, and Peña, J M In 2017 IEEE Congress on Evolutionary Computation (CEC 2017) 2017 [Abs] [HTML]

   The scalability of optimization algorithms is an important issue that has been thoroughly studied in the past. However, these studies were normally conducted by gradually increasing the dimensionality of the benchmark and analyzing how an algorithm exhibiting a good performance on low-dimensional problems degrades as the problem size increases. In this contribution we follow the opposite approach: we take some well-known large-scale global optimizers based on the MOS framework and specifically designed for problems of thousands of variables and evaluate them on much smaller problems (up to 100 dimensions). The results show that, surprisingly, these algorithms are able to find good solutions to many of the functions of the benchmark, systematically reaching the global optimum for some of them. Furthermore, the differences in performance among the three considered algorithms are also analyzed and compared with statistical methods. Finally, several hypothesis are given to explain these differences in performance among the three algorithms.

2015

1. Comparative Study of Metaheuristics for the Curve-Fitting Problem: Modeling Neurotransmitter Diffusion and Synaptic Receptor Activation Montes, J, LaTorre, A, Muelas, S, Merchán-Pérez, A, and Peña, J M Abstract and Applied Analysis 2015 [Abs] [HTML]

   Synapses are key elements in the information transmission in the nervous system. Among the different approaches to study them, the use of computational simulations is identified as the most promising technique. Simulations, however, do not provide generalized models of the underlying biochemical phenomena, but a set of observations, or time-series curves, displaying the behavior of the synapse in the scenario represented. Finding a general model of these curves, like a set of mathematical equations, could be an achievement in the study of synaptic behavior. In this paper, we propose an exploratory analysis in which selected curve models are proposed, and state-of-the-art metaheuristics are used and compared to fit the free coefficients of these curves to the data obtained from simulations. Experimental results demonstrate that several models can fit these data, though a deeper analysis from a biological perspective reveals that some are better suited for this purpose, as they represent more accurately the biological process. Based on the results of this analysis, we propose a set of mathematical equations and a methodology, adequate for modeling several aspects of biochemical synaptic behavior.

2. A Distributed VNS Algorithm for Optimizing Dial-a-Ride Problems in Large-Scale Scenarios Muelas, S, LaTorre, A, and Peña, J M Transportation Research Part C: Emerging Technologies 2015 [Abs] [HTML]

   These days, transportation and logistic problems in large cities are demanding smarter transportation services that provide flexibility and adaptability. A possible solution to this arising problem is to compute the best routes for each new scenario. In this problem, known in the literature as the dial-a-ride problem, a number of passengers are transported between pickup and delivery locations trying to minimize the routing costs while respecting a set of prespecified constraints. This problem has been solved in the literature with several approaches from small to medium sized problems. However, few e\carriagereturnorts have dealt with large scale scenarios very common in massive scenarios (big cities or high-populated regions). In this study, a new distributed algorithm based on the partition of the requests space and the combination of the routes is presented and tested on a set of 24 di\carriagereturnerent scenarios of a large-scale problem (up to 16,000 requests or 32,000 locations) in the city of San Francisco. The results show that, not only the distributed algorithm is able to solve large problem instances that the corresponding sequential algorithm is unable to solve in a reasonable time, but also to have an average improvement of 9% in the smaller problems. The results have been validated by means of statistical procedures proving that the distributed algorithm can be an e\carriagereturnective way to solve high dimensional dial-a-ride problems.

2014

1. Designer-Driven 3D Buildings Generated Using Variable Neighborhood Search Peña, J M, Viedma, J, Muelas, S, LaTorre, A, and Peña, L In 2014 IEEE Conference on Computational Intelligence and Games (CIG 2014) 2014 [Abs]

   This paper presents a mechanism to generate virtual buildings, for action or RPG games, considering designer constrains and guidelines. This mechanism is implemented as a pipeline of different Variable Neighborhood Search (VNS)optimization processes in which several subproblems are tackled (1) rooms locations, (2) connectivity graph, and (3) element placement. The core VNS algorithm includes some variants to improve its performance, such as, for example constrain handling and biased operator selection. The optimization process uses a toolkit of construction primitives implemented as “smart objects” providing basic elements such as rooms, doors, staircases and other connectors. The paper also shows experimental results of the application of different designer constrains to a wide range of buildings from small houses to a large castle with several underground levels.

2. Distributed Estimation of Distribution Algorithms for Continuous Optimization: How Does the Exchanged Information Influence Their Behavior? Muelas, Santiago, Mendiburu, Alexander, LaTorre, A, and Peña, J M Information Sciences 2014 [Abs] [HTML]

   One of the most promising areas in which probabilistic graphical models have shown an incipient activity is the field of heuristic optimization and, in particular, in Estimation of Distribution Algorithms. Due to their inherent parallelism, different research lines have been studied trying to improve Estimation of Distribution Algorithms from the point of view of execution time and/or accuracy. Among these proposals, we focus on the so-called distributed or island-based models. This approach defines several islands (algorithms instances) running independently and exchanging information with a given frequency. The information sent by the islands can be either a set of individuals or a probabilistic model. This paper presents a comparative study for a distributed univariate Estimation of Distribution Algorithm and a multivariate version, paying special attention to the comparison of two alternative methods for exchanging information, over a wide set of parameters and problems – the standard benchmark developed for the IEEE Workshop on Evolutionary Algorithms and other Metaheuristics for Continuous Optimization Problems of the ISDA 2009 Conference. Several analyses from different points of view have been conducted to analyze both the influence of the parameters and the relationships between them including a characterization of the configurations according to their behavior on the proposed benchmark.

3. A Comprehensive Comparison of Large Scale Global Optimizers LaTorre, A, Muelas, S, and Peña, J M Information Sciences 2014 [Abs] [HTML]

   Large Scale Global Optimization is one of the most active research lines in evolution- ary and metaheuristic algorithms. In the last five years, several conference sessions and journal special issues have been conducted, and many algorithmic alternatives and hy- brid methods, more and more sophisticated, have been proposed. However, most of the proposed algorithms are only evaluated on a particular benchmark of functions and thus its performance in other benchmarks presenting different characteristics remains unknown. In this paper, it is our aim to fill in this gap by evaluating and comparing 10 of the most recently proposed algorithms, in particular, those reporting the best per- formance in the last major competitions. This paper proposes an evaluation consisting of a broader testbed that considers all the functions of three well-known benchmarks, including a comparative statistical study of the results and the identification of algo- rithm profiles for those with an equivalent performance. As a part of the comparative analysis this paper also includes three different studies: (1) first, on the complexity of the compared algorithms; (2) then, on the relevance of the comparative statistical tests; and (3) finally, on direct/indirect measures of the exploration/exploitation capabilities of the most representative algorithms in the overall comparison. In addition, this work introduces an open-access web service to perform future analysis and keep trace of new algorithm performances offered to the community of researchers in the field.

2013

1. Large Scale Global Optimization: Experimental Results with MOS-Based Hybrid Algorithms LaTorre, A, Muelas, S, and Peña, J M In 2013 IEEE Congress on Evolutionary Computation (CEC 2013) 2013
2. Evaluating the Multiple Offspring Sampling Framework on Complex Continuous Optimization Functions LaTorre, A, Muelas, S, and Peña, J M Memetic Computing 2013 [Abs] [HTML]

   In this contribution we present a study on the combination of Dif- ferential Evolution and the IPOP-CMA-ES algorithms. The hybrid algorithm has been constructed by using the Multiple Offspring Sampling framework, which allows the seamless combination of multiple metaheuristics in a dy- namic algorithm capable of adjusting the participation of each of the com- posing algorithms according to their current performance. In this study we analyze the existing synergies, if any, emerging from the combination of the two algorithms. For this purpose, the COCO suite used in BBOB 2009 and 2010 Workshops has been used. The experimental results on the noiseless testbed show a robust behavior of the algorithm and a good scalability as the dimensionality increases. In the noisy testbed, the algorithm shows a good performance on functions with moderate to severe noise.

3. Soft Computing for Content Generation: Trading Market in a Basketball Management Video Game Peña, J M, Menasalvas, E, Muelas, S, LaTorre, A, Peña, L, and Ossowski, S In 2013 IEEE Conference on Computational Intelligence and Games (CIG 2013) 2013 [Abs]

   Although procedural and assisted content gener- ation have attracted a lot of attention in both academic and industrial research in video games, there are few cases in the literature in which they have been applied to sport management games. The on-line variants of these games produce a lot of information concerning how the users interact with each other in the game. This contribution presents the application of soft computing techniques in the context of content generation for an on-line massive basketball management simulation game (in particular in the virtual trading market of the game). This application is developed in two different directions: (1) a machine learning model to analyze the appeal of the trading market contents (the virtual basketball players in the game), and (2) an evolutionary algorithm to assist users in the design of new contents (training of virtual basketball players).

4. Segmentation of Neuronal Nuclei Based on Clump Splitting and a Two-Step Binarization of Images LaTorre, A, Alonso-Nanclares, L, Muelas, S, Peña, J M, and DeFelipe, J Expert Systems with Applications 2013 [Abs] [HTML]

   In this paper we present an algorithm to segment the nuclei of neuronal cells in confocal microscopy images, a key technical problem in many experimental studies in the field of neuroscience. We describe the whole procedure, from the original images to the segmented individual nuclei, paying particular attention to the binarization of the images, which is not straightforward due to the technical difficulties related to the visualization of nuclei as individual objects and incomplete and irregular staining. We have focused on the division of clusters of nuclei that appear frequently in these images. Thus we have devel- oped a clump-splitting algorithm to separate touching or overlapping nuclei allowing us to accurate account for both the number and size of the nuclei. The results presented in the paper show that the proposed algorithm performs well on different sets of images from different layers of the cerebral cortex.

5. A Variable Neighborhood Search Algorithm for the Optimization of a Dial-a-Ride Problem in a Large City Muelas, Santiago, LaTorre, A, and Peña, J M Expert Systems with Applications 2013 [Abs] [HTML]

   On-demand transportation is becoming a new necessary service for modern (public and private) mobility and logistics providers. Large cities are demanding more and more share transportation services with flexible routes, resulting from user dynamic demands. In this study a new algorithm is proposed for solving the problem of computing the best routes that a public transportation company could offer to satisfy a number of customer requests. In this problem, known in the literature as the Dial-a-ride problem, a number of passengers has to be transported between pickup and delivery locations trying to minimize the routing costs while respecting a set of pre-specified constraints (maximum pickup time, maximum ride duration and maximum load per vehicle). For optimizing this problem, a new Variable neighborhood search has been developed and tested on a set of 24 different scenarios of a large-scale Dial-a-ride problem in the city of San Francisco. The results have been compared against two state-of-the-art algorithms of the literature and validated by means of statistical procedures proving that the new algorithm has obtained the best overall results.

6. 3D Segmentations of Neuronal Nuclei from Confocal Microscope Image Stacks LaTorre, A, Alonso-Nanclares, Lidia, Muelas, Santiago, Peña, J M, and DeFelipe, J Frontiers in Neuroanatomy 2013 [Abs] [HTML]

   In this paper, we present an algorithm to create 3D segmentations of neuronal cells from stacks of previously segmented 2D images. The idea behind this proposal is to provide a general method to reconstruct 3D structures from 2D stacks, regardless of how these 2D stacks have been obtained. The algorithm not only reuses the information obtained in the 2D segmentation, but also attempts to correct some typical mistakes made by the 2D segmentation algorithms (for example, under segmentation of tightly-coupled clusters of cells). We have tested our algorithm in a real scenario — the segmentation of the neuronal nuclei in different layers of the rat cerebral cortex. Several representative images from different layers of the cerebral cortex have been considered and several 2D segmentation algorithms have been compared. Furthermore, the algorithm has also been compared with the traditional 3D Watershed algorithm and the results obtained here show better performance in terms of correctly identified neuronal nuclei.

2012

1. Multiple Offspring Sampling in Large Scale Global Optimization LaTorre, A, Muelas, S, and Peña, J M In 2012 IEEE Congress on Evolutionary Computation (CEC 2012) 2012 [Abs]

   Continuous optimization is one of the most active research lines in evolutionary and metaheuristic algorithms. Through CEC 2005 to CEC 2011 competitions, many different algorithms have been proposed to solve continuous problems. The advances on this type of problems are of capital importance as many real-world problems from very different domains (biology, engineering, data mining, etc.) can be formulated as the optimization of a continuous function. In this paper we analyse the behaviour of a hybrid algorithm combining two heuristics that have been successfully applied to solving continuous optimization problems in the past. We show that the combination of both algorithms obtains competitive results on the proposed benchmark by automatically selecting the most appropriate heuristic for each function and search phase.

2. Detecting Synaptic Vesicles by Means of a Differential Evolution Algorithm LaTorre, A, Muelas, S, and Peña, J M In VIII Congreso Español Sobre Metaheurísticas, Algoritmos Evolutivos y Bioinspirados (MAEB 2012) 2012 [Abs] [HTML]

   Neurotransmitters used by chemical sy- napses are stored in synaptic vesicles that accumu- late in axon terminals. The number and position of these vesicles have been related to some important functional properties of the synapse. For this reason, an accurate mechanism for semi-automatically coun- ting these small cellular structures will be of great help for neuroscientists. In this paper, we present a Differential Evolution algorithm that quantifies the number of synaptic vesicles in electron micrographs. The algorithm has been tested on several images that have been obtained from the somatosensory cortex of the rat and compared with some traditional approa- ches for detecting circular structures. Finally, the re- sults have been validated by two independent expert anatomists.

3. A New Methodology for the Automatic Creation of Adaptive Hybrid Algorithms Muelas, S, LaTorre, A, and Peña, J M Intelligent Data Analysis 2012 [Abs] [HTML]

   Hybrid Evolutionary Algorithms are a promising alternative to deal with the problem of selecting the most appropriate Evolutionary Algorithm for a specific problem. By means of the combination of different heuristic optimization ap- proaches, it is possible to profit from the benefits of the best approach or, even more, to discover synergies between the algorithms that could improve the results of the best performing individual algorithm. Nowadays, there is an active research in the design of dynamic or adaptive combination strategies for hybrid algorithms. However, little research has been done in the automatic learning of the best hy- bridization strategy. This paper proposes a new methodology for developing in- telligent adaptive hybrid algorithms that uses data mining techniques to analyze the results from past executions. The proposed methodology has been evaluated on a well-known benchmark on continuous optimization made up of 19 differ- ent functions and several dimensions (50, 100, 200 and 500). Several analyses have been conducted and statistical tests have been used for validating the results. The generated hybrid algorithm has achieved outstanding results, obtaining signif- icantly better results than the MOS algorithm, the most performant algorithm on this benchmark, and the CMA-ES algorithm, one of the reference algorithms in continuous optimization.1

2011

1. A New Initialization Procedure for the Distributed Estimation of Distribution Algorithms Muelas, S, Peña, J M, LaTorre, A, and Robles, V Soft Computing - A Fusion of Foundations, Methodologies and Applications 2011 [Abs] [HTML]

   Estimation of Distribution Algorithms (EDAs) are one of the most promising paradigms in today’s evolutionary computation. In this field there has been an incipient activity in the so called parallel Estimation of Distribution Algorithms (pEDAs). One of these approaches is the distributed Estimation of Distribution Algorithms (dEDAs). This paper introduces a new initialization mechanism for each of the populations of the islands based on Voronoi cells. In order to analyze the results, a series of different experiments using the benchmark suite for the special session on Real-parameter Optimization of the IEEE CEC 2005 conference has been carried out. The results obtained suggest that the Voronoi initialization method considerably improves the performance obtained from a traditional uniform initialization.

2. Benchmarking a Hybrid DE-RHC Algorithm on Real World Problems LaTorre, A, Muelas, S, and Peña, J M In 2011 IEEE Congress on Evolutionary Computation (CEC 2011) 2011 [Abs] [HTML]

   Continuous optimization is one of the most active research lines in evolutionary and metaheuristic algorithms. Through CEC 2005 to CEC 2010 competitions, many different algorithms have been proposed to solve continuous problems. The advances on this type of problems are of capital importance as many real-world problems from very different domains (biology, engineering, data mining, etc.) can be formulated as the optimization of a continuous function. For this reason, we have proposed a hybrid DE-RHC algorithm that combines the search strength of Differential Evolution with the explorative ability of a Random Hill Climber, which can help the Differential Evolution algorithm to reach new promising areas in difficult fitness landscapes, such as those than can be found on real-world problems. To evaluate this approach, the benchmark problems proposed in the "Testing Evolutionary Algorithms on Real-world Numerical Optimization Problems" CEC 2011 special session have been considered.

3. A Differential Evolution Algorithm for the Detection of Synaptic Vesicles LaTorre, A, Muelas, S, Peña, J M, Santana, R, Merchán-Pérez, A, and Rodriguez, J R In 2011 IEEE Congress on Evolutionary Computation (CEC 2011) 2011 [Abs] [HTML]

   Neurotransmitters used by chemical synapses are stored in synaptic vesicles that accumulate in axon terminals. The number and position of these vesicles have been related to some important functional properties of the synapse. For this reason, an accurate mechanism for semi-automatically counting these small cellular structures will be of great help for neuroscientists. In this paper, we present a Differential Evolution algorithm that quantifies the number of synaptic vesicles in electron micrographs. The algorithm has been tested on several images that have been obtained from the somatosensory cortex of the rat and compared with some traditional approaches for detecting circular structures. Finally, the results have been validated by two independent expert anatomists.

4. A Direct Optimization Approach to the P300 Speller Santana, R, Muelas, S, LaTorre, A, and Peña, J M In 13th Genetic and Evolutionary Computation Conference (GECCO 2011) 2011 [Abs] [HTML]

   The P300 component of the brain event-related-potential is one of the most used signals in brain computer interfaces (BCIs). One of the required steps for the application of the P300 paradigm is the identification of this component in the presence of stimuli. In this paper we propose a direct optimization approach to the P300 classification problem. A general formulation of the problem is introduced. Different classes of optimization algorithms are applied to solve the problem and the concepts of k-best and k-worst ensembles of solutions are introduced as a way to improve the accuracy of single solutions. The introduced approaches are able to achieve a classification rate over 80% on test data.

2010

1. On the Evaluation of Information Exchange Strategies in dEDAs Muelas, S, Mendiburu, A, LaTorre, A, and Peña, J M In VII Congreso Español Sobre Metaheurísticas, Algoritmos Evolutivos y Bioinspirados (MAEB 2010) 2010 [Abs]

   One of the most promising areas in which probabilistic graphical models have shown an incipient activity is the field of heuristic opti- mization and, in particular, in the Estimation of Distribution Algorithms (EDAs). EDAs constitute a well-known family of Evolutionary Computation techniques, similar to Genetic Algorithms. Due to their inherent parallelism, different research lines have tried to improve EDAs from the point of view of execution time and/or accuracy. Among these proposals, we focus on the so-called island-based models. This approach defines several islands (EDA in- stances) running independently and exchang- ing information with a given frequency. The information sent by the islands can be a set of individuals or a probabilistic model. This pa- per presents a comparative study of both infor- mation exchanging techniques for a univariate EDA (UMDAg) over a wide set of parame- ters and problems –the standard benchmark developed for the IEEE Workshop on Evolu- tionary Algorithms and other Metaheuristics for Continuous Optimization Problems of the ISDA 2009 Conference.

2. La Seguridad En Arquitecturas: Virtualización, Grid y Cloud Computing and LaTorre, A 2010
3. Learning Hybridization Strategies in Evolutionary Algorithms LaTorre, A, Peña, J M, Muelas, S, and Freitas, A A Intelligent Data Analysis 2010 [Abs] [HTML]

   Evolutionary Algorithms are powerful optimization techniques which have been applied to many different problems, from complex mathematical functions to real-world applications. Some studies report performance improvements through the combination of different evolutionary approaches within the same hybrid algorithm. However, the mechanisms used to control this combination of evolutionary approaches are not as satisfactory as would be desirable. In most cases, there is no feedback from the algorithm nor any regulatory component that modifies the participation of each evolutionary approach in the overall search process. In some cases, the algorithm makes use of some information for an on-line adaptation of the participation of each algorithm. In this paper, the use of Reinforcement Learning (RL) is proposed as a mechanism to control how the different evolutionary approaches contribute to the overall search process. In particular, three learning policies based on one of the state-of-the-art RL algorithms, Q-Learning, have been considered and used to control the participation of each algorithm by learning the best-response mixed strategy. To test this approach, a benchmark made up of six large-scale (500 dimensions) continuous optimization functions has been considered. The experimentation carried out has proved that RL control mechanisms successfully learn optimal patterns for the combination of evolutionary algorithms in most of the proposed functions, being able to improve the performance of both individual and non RL hybrid algorithms.

4. MOS LaTorre, A, Peña, J M, and Zaforas, M Matematicalia 2010 [Abs] [HTML]

   MOS. Dentro del campo de la inteligencia artificial, los algoritmos evolutivos (que se inspiran en la teoría de la evolución postulada por Darwin) permiten obtener soluciones a problemas de búsqueda y optimización no lineales y muy complejos. J.M. Peña, A. LaTorre y M. Zaforas (Universidad Politécnica de Madrid) presentan a nuestros lectores el Framework MOS para el desarrollo de algoritmos evolutivos híbridos con ajuste dinámico de participación.

5. La Seguridad En Los Sistemas Operativos: Los Sistemas Operativos (ESA, UNIX, Windows, Linux) LaTorre, A, Robles, V, and Pérez, M S 2010
6. A MOS-Based Dynamic Memetic Differential Evolution Algorithm for Continuous Optimization: A Scalability Test LaTorre, A, Muelas, Santiago, and Peña, J M Soft Computing - A Fusion of Foundations, Methodologies and Applications 2010 [Abs] [HTML]

   Continuous optimization is one of the areas with more activity in the field of heuristic optimization. Many algorithms have been proposed and compared on several benchmarks of functions, with different perfor- mance depending on the problems. For this reason, the combination of different search strategies seems desirable to obtain the best performance of each of these approaches. This contribution explores the use of a hybrid memetic algorithm based on the multiple offspring framework. The proposed algorithm combines the explorative/exploitative strength of two heuristic search methods that separately obtain very competitive results. This algorithm has been tested with the benchmark problems and conditions defined for the special issue of the Soft Computing Journal on Scalability of Evolutionary Algorithms and other Meta- heuristics for Large Scale Continuous Optimization Prob- lems. The proposed algorithm obtained the best results compared with both its composing algorithms and a set of reference algorithms that were proposed for the special issue.

7. An Analysis of a Hybrid Evolutionary Algorithm by Means of Its Phylogenetic Information LaTorre, A, Muelas, S, and Peña, J M In 2010 World Automation Congress (WAC 2010) 2010 [Abs] [HTML]

   The study conducted in this work analyses the interactions between different Evolutionary Algorithms when they are hybridized. For this purpose, the phylogenetic tree of the best solution reported by the hybrid algorithm is reconstructed, and the relationships among the ancestors of this solution are established. For each of these ancestors, the evolutionary techniques that generated that solution and the fitness increment introduced compared to its parents are recorded. The study reveals a structured interaction among the different evolutionary techniques that makes the hybrid algorithm to outperform each of its composing algorithms when executed individually. The Multiple Offspring Sampling framework has been used to develop the Hybrid EA studied in this work and the experiments have been conducted on the well-known CEC 2005 Benchmark for continuous optimization.

8. Benchmarking a MOS-Based Algorithm on the BBOB-2010 Noisy Function Testbed LaTorre, A, Muelas, S, and Peña, J M In 12th Genetic and Evolutionary Computation Conference (GECCO 2010) 2010 [Abs] [HTML]

   In this paper, a hybrid algorithm based on the Multiple Offspring Sampling framework is presented and benchmarked on the BBOB-2010 noisy testbed. MOS allows the seamless combination of multiple metaheuristics in a hybrid algorithm capable of dynamically adjusting the participation of each of the composing algorithms. The experimental results show a good performance on functions with moderate noise. However, on functions with severe noise the results deteriorate, which suggests that further research should be conducted to find more adequate control mechanisms for these types of functions.

9. Migrating Individuals and Probabilistic Models on dEDAs: A Comparison on Continuous Functions Muelas, S, Mendiburu, A, LaTorre, A, and Peña, J M In 10th IASTED International Conference on Artificial Intelligence and Applications (AIA 2010) 2010 [Abs]

   One of the most promising areas in which probabilistic graphical models have shown an incipient activity is the field of heuristic optimization and, in particular, in the Estimation of Distribution Algorithms (EDAs). EDAs constitute a well-known family of Evolutionary Computation techniques, similar to Genetic Algorithms. Due to their inherent parallelism, different research lines have been studied trying to improve EDAs from the point of view of execution time and/or accuracy. Among these proposals, we focus on the so-called island-based models. This approach defines several islands (EDA instances) running independently and exchanging information with a given frequency. The information sent by the islands can be a set of individuals or a probabilistic model. This paper presents a comparative study of both information exchanging techniques for a univariate EDA (\UMDA_g\) over a wide set of parameters and problems –the standard benchmark developed for the IEEE Workshop on Evolutionary Algorithms and other Metaheuristics for Continuous Optimization Problems of the ISDA 2009 Conference. The study concludes that the configurations based on migrating individuals obtain better results.

10. Automatically Modeling Hybrid Evolutionary Algorithms from Past Executions Muelas, S, Peña, J M, and LaTorre, A In Applications of Evolutionary Computing: EvoWorkshops 2010: EvoCOMNET, EvoENVIRONMENT, EvoFIN, EvoGAMES, EvoHOT, EvoIASP, EvoINTERACTION, EvoMUSART, EvoNUM, EvoSTOC, EvoTRANSLOG 2010 [Abs] [HTML]

    The selection of the most appropriate Evolutionary Algorithm for a given optimization problem is a difficult task. Hybrid Evolutionary Algorithms are a promising alternative to deal with this problem. By means of the combination of different heuristic optimization approaches, it is possible to profit from the benefits of the best approach, avoiding the limitations of the others. Nowadays, there is an active research in the design of dynamic or adaptive hybrid algorithms. However, little research has been done in the automatic learning of the best hybridization strategy. This paper proposes a mechanism to learn a strategy based on the analysis of the results from past executions. The proposed algorithm has been evaluated on a well-known benchmark on continuous optimization. The obtained results suggest that the proposed approach is able to learn very promising hybridization strategies.

11. Benchmarking a MOS-Based Algorithm on the BBOB-2010 Noiseless Function Testbed LaTorre, A, Muelas, S, and Peña, J M In 12th Genetic and Evolutionary Computation Conference (GECCO 2010) 2010 [Abs] [HTML]

    In this contribution, a hybrid algorithm combining Differential Evolution and IPOP-CMA-ES is presented and benchmarked on the BBOB 2010 noiseless testbed. The hybrid algorithm has been constructed within the Multiple Offspring Sampling framework, which allows the seamless combination of multiple metaheuristics in a dynamic algorithm capable of adjusting the participation of each of the composing algorithms according to their current performance. The experimental results show a robust behavior of the algorithm and a good scalability as the dimensionality increases.

2009

1. A Framework for Hybrid Dynamic Evolutionary Algorithms: Multiple Offspring Sampling (MOS) and LaTorre, A 2009
2. Algoritmos Distribuidos Heterogéneos Para Problemas de Optimización Continua Muelas, S, Peña, J M, LaTorre, A, and Robles, V In VI Congreso Español Sobre Metaheurísticas, Algoritmos Evolutivos y Bioinspirados (MAEB 2009) 2009
3. MOS Como Herramienta Para La Hibridación de Algoritmos Evolutivos LaTorre, A, Peña, J M, Fernández, J, and Muelas, S In VI Congreso Español Sobre Metaheurísticas, Algoritmos Evolutivos y Bioinspirados (MAEB 2009) 2009
4. Quality Measures to Adapt the Participation in MOS LaTorre, A, Peña, J M, Muelas, S, and Pascual, C In 2009 IEEE Congress on Evolutionary Computation (CEC 2009) 2009
5. Tentative Exploration on Reinforcement Learning Algorithms for Stochastic Rewards Peña, L, LaTorre, A, Peña, J M, and Ossowski, S In 4th International Conference on Hybrid Artificial Intelligence Systems (HAIS 2009) 2009
6. Hybrid Evolutionary Algorithms for Large Scale Continuous Problems LaTorre, A, Peña, J M, Muelas, S, and Zaforas, M In 11th Genetic and Evolutionary Computation Conference (GECCO 2009) 2009
7. A Memetic Differential Evolution Algorithm for Continuous Optimization Muelas, S, LaTorre, A, and Peña, J M In 9th International Conference on Intelligent Systems Design and Applications (ISDA 2009) 2009 [Abs] [HTML]

   Continuous optimization is one of the most active research lines in evolutionary and metaheuristic algorithms. Since CEC 2005 and CEC 2008 competitions, many different algorithms have been proposed to solve continuous problems. Despite there exist very good algorithms reporting high quality results for a given dimension, the scalability of the search methods is still an open issue. Finding an algorithm with competitive results in the range of 50 to 500 dimensions is a difficult achievement. This contribution explores the use of a hybrid memetic algorithm based on the differential evolution algorithm, named MDE-DC. The proposed algorithm combines the explorative/exploitative strength of two heuristic search methods, that separately obtain very competitive results in either low or high dimensional problems. This paper uses the benchmark problems and conditions required for the workshop on “evolutionary algorithms and other metaheuristics for Continuous Optimization Problems – A Scalability Test” chaired by Francisco Herrera and Manuel Lozano.

2008

1. VRP-Extended: When Confidence and Fleet Size Are Also Important LaTorre, A, Clautiaux, F, Talbi, E-G, and Peña, J M In 2nd International Conference on Metaheuristics and Nature Inspired Computing (META 2008) 2008 [Abs]

   In this paper we present an extension to the traditional Vehicle Routing Problem where two new constraints have been considered: a set of fixed customers that must be visited altogether and a limit on the available number of vehicles that forces the delivery company to subcontract some customers to another company. A genetic algorithm has been designed and successfully applied to solve this new problem.

2. Using Multiple Offspring Sampling to Guide Genetic Algorithms to Solve Permutation Problems LaTorre, A, Peña, J M, Robles, V, and Muelas, S In 10th Genetic and Evolutionary Computation Conference (GECCO 2008) 2008 [Abs] [HTML]

   The correct choice of an evolutionary algorithm, a genetic representation for the problem being solved (as well as their associated variation operators) and the appropriate values for the parameters of the algorithm is a hard task and it is often considered as an optimization problem itself. In this contribution, we propose a new theoretical formalism, called Multiple Offspring Sampling (MOS). This new technique combines different evolutionary approaches taking advantage of the benefits provided by each of them. MOS dynamically balances the participation of different mechanisms to spawn the new offspring population, according to the benefits provided by each of them in previous generations. This approach evaluates multiple offspring generation methods (for example different coding strategies), and configures appropriate sampling sizes. This formalism has been applied to a well-known permutation problem, the traveling salesman problem (TSP). The results on several instances of this problem show that most of the combined techniques outperform the results obtained by single ones.

3. Supercomputer Scheduling with Combined Evolutionary Techniques LaTorre, A, Peña, J M, Robles, V, and de Miguel, P 2008 [Abs] [HTML]

   Schedulingisaveryimportantprobleminmanyreal-worldscenarios.Inthecase of supercomputers it is even more important because available resources are limited and ex- pensive. The optimal use of supercomputer facilities is a critical question. We have found that the definitions of traditional scheduling problems do not provide an appropriate description for Supercomputer Scheduling (SCS). Thus, a new definition for this kind of problems is pro- posed. The research already done in the field of other scheduling problems can be modified to be applied in this new scenario. Nevertheless, new techniques can also be developed; we have proposed a theoretical framework to combine multiple evolutionary algorithms called Multiple Offspring Sampling (MOS). We have used this approach to combine multiple cod- ings and genetic operators in this scheduling problem. To summarise: first, we introduce a formal definition of supercomputer scheduling; second, we propose Multiple Offspring Sam- pling formalism; and third, we have carried out an experimental test to compare the usage of this formalism to solve SCS problem against traditional (non-combinatorial) techniques and single genetic algorithms.

4. Optimizing the Design of Composite Panels Using an Improved Genetic Algorithm Muelas, S, Peña, J M, Robles, V, Muzhetskaya, K, LaTorre, A, and de Miguel, P In 2008 International Conference on Engineering Optimization (EngOpt 2008) 2008 [Abs]

   Composite materials are combinations of, at least, two organic or inorganic materials, working to- gether to give the composite some desired properties. Composites are highly-used on industrial design [3]. Their light weight make them key elements to reduce weight and direct operating costs in some domains like aeronautics. Genetic algorithms are heuristic stochastic methods that explore a reduced set of tentative solutions, performing a guided search procedure that evaluates few solutions, in several orders of magnitude smaller than the whole search space. The dynamics in genetic algorithms provide optimal (or near-optimal) solutions to complex optimization problems when analytical techniques are not able to calculate them. This paper describes the results of applying several improvements to the standard genetic algorithm to the optimization of a stiffened composite panel sub ject to a set of shear and axial loads. The performance and characteristics of the proposed configurations are evaluated via nonlinear finite element simulation. The goal is to find the lightest configuration that keeps the principal strains under a given threshold. The proposed improvements significantly reduce both the weight and the number of analyses required for the optimization.

5. Voronoi-Initializated Island Models for Solving Real-Coded Deceptive Problems Muelas, S, Peña, J M, Robles, V, and LaTorre, A In 10th Genetic and Evolutionary Computation Conference (GECCO 2008) 2008 [Abs] [HTML]

   Deceptive problems have always been considered difficult for Genetic Algorithms. To cope with this characteristic, the literature has proposed the use of Parallel Genetic Algorithms (PGAs), particularly multi-population island-based models. Although the existence of multiple populations encourages population diversity, these problems are still difficult to solve. This paper introduces a new initialization mechanism for each of the populations of the islands based on Voronoi cells. In order to analyze the results, a series of different experiments using several real-value deceptive problems and a set of representative parameters (migration ratio, migration frequency and connectivity) have been chosen. The results obtained suggest that the Voronoi initialization method improves considerably the performance obtained with a traditionally uniform random initialization.

2007

1. Machine Learning Methods to Analyze Migration Parameters in Parallel Genetic Algorithms Muelas, S, Peña, J M, Robles, V, LaTorre, A, and de Miguel, P In 2nd International Workshop on Hybrid Artificial Intelligence Systems (HAIS 2007) 2007 [Abs] [HTML]

   Parallel genetic algorithms (PGAs) are a powerful tool to deal with complex optimization problems. Nevertheless, the task of selecting its parameters accurately is an optimization problem by itself. Any additional help or hints to adjust the configuration parameters will lead both towards a more efficient PGA application and to a better comprehension on how these parameters affect optimization behavior and performance. This contribution offers an analysis on certain PGA parameters such as migration frequency, topology, connectivity and number of islands. The study has been carried out on an intensive set of experiments that collect PGA performance on several representative problems. The results have been analyzed using machine learning methods to identify behavioral patterns that are labeled as “good” PGA configurations. This study is a first step to generalize relevant patterns from the problems analyzed that identify better configurations in PGAs.

2. Breast Cancer Biomarker Selection Using Multiple Offspring Sampling LaTorre, A, Peña, J M, González, S, Robles, V, and Famili, F In 27th European Conference on Machine Learning and 20th Principles & Practice of Knowledge Discovery (ECML/PKDD 2007): Workshop on Data Mining in Functional Genomics and Proteomics: Current Trends and Future Directions 2007

2005

1. Heuristic Optimization Techniques Cubo, O, LaTorre, A, Robles, V, Pérez, M S, González, S, and Peña, J M 2005