S2 Grupo

Valencia, Spain
Valencia, Spain
SEARCH FILTERS
Time filter
Source Type

Merelo J.J.,University of Granada | Castillo P.,University of Granada | Mora A.,University of Granada | Esparcia-Alcazar A.I.,S2 Grupo
GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference | Year: 2013

Solving the MasterMind puzzle, that is, finding out a hidden combination by using hints that tell you how close some strings are to that one is a combinatorial optimization problem that becomes increasingly difficult with string size and the number of symbols used in it. Since it does not have an exact solution, heuristic methods have been traditionally used to solve it; these methods scored each combination using a heuristic function that depends on comparing all possible solutions with each other. In this paper we first optimize the implementation of previous evolutionary methods used for the game of mastermind, obtaining up to a 40% speed improvement over them. Then we study the behavior of an entropy-based score, which has previously been used but not checked exhaustively and compared with previous solutions. The combination of these two strategies obtain solutions to the game of Mastermind that are competitive, and in some cases beat, the best solutions obtained so far. All data and programs have also been published under an open source license. Copyright © 2013 ACM.


Merelo J.J.,University of Granada | Mora A.M.,University of Granada | Fernandes C.M.,University of Granada | Esparcia-Alcazar A.I.,S2 Grupo
Natural Computing | Year: 2013

This paper focuses on the iterative design of SofEA, an architecture for distributing evolutionary algorithms (EAs) across computer networks in an asynchronous and decentralized way. SofEA is based on a pool architecture implemented on an object store, allowing the asynchronous interaction with which several clients. The fact that each client is autonomous leads to complex behavior, which will be examined in the work, so that the design can be validated, rules of thumb can be extracted, and the limits of scalability can be found. In this paper we advance the design of an asynchronous, fault-tolerant, and hopefully scalable distributed EA based on the object store CouchDB. We do so by iteratively analyzing running time and average evaluations to solutions on increasingly better versions of the algorithm, looking for the best results, at least from the point of view of running time. By doing so, we increase speed almost fourfold, and also decrement the average number of evaluations to solution in some cases. Experiments have shown also which critical parameters have the bigger influence on the performance in this kind of systems: live population size and number of conflicts, with both being influenced by the number of clients and the size of the population block each client handles at a time. These experiments also show that there is a balance between scalability and fault tolerance, with scalability dropping when a certain number of clients is reached; further clients only increase fault tolerance, at least in the configurations we are using in this paper. The paper also shows that experimentation and measurement conform a good methodology for the design of this kind of asynchronous, heterogeneous and distributed systems, where analytic performance prediction is almost impossible. © 2012 Springer Science+Business Media B.V.


Mora A.M.,University of Granada | De Las Cuevas P.,University of Granada | Merelo J.J.,University of Granada | Zamarripa S.,S2 Grupo | Esparcia-Alcazar A.I.,S2 Grupo
GECCO 2014 - Companion Publication of the 2014 Genetic and Evolutionary Computation Conference | Year: 2014

This paper presents an approach, based in a project in development, which combines Data Mining, Machine Learning and Computational Intelligence techniques, in order to create a user-centric and adaptable corporate security system. Thus, the system, named MUSES, will be able to analyse the user's behaviour (modelled as events) when interacting with the company's server, accessing to corporate assets, for instance. As a result of this analysis, and after the application of the aforementioned techniques, the Corporate Security Policies, and specifically, the Corporate Security Rules will be adapted to deal with new anomalous situations, or to better manage user's behaviour. The work reviews the current state of the art in security issues resolution by means of these kind of methods. Then it describes the MUSES features in this respect and compares them with the existing approaches.


Merelo J.J.,University of Granada | Mora A.M.,University of Granada | Fernandes C.M.,University of Granada | Esparcia-Alcazar A.I.,S2 Grupo | Laredo J.L.J.,University of Luxembourg
Proceedings - 2012 7th International Conference on P2P, Parallel, Grid, Cloud and Internet Computing, 3PGCIC 2012 | Year: 2012

This paper explores the scalability and performance of pool and island based evolutionary algorithms, both of them using as a mean of interaction an object store, we call this family of algorithms SofEA. This object store allows the different clients to interact asynchronously, the point of the creation of this framework is to build a system for spontaneous and voluntary distributed evolutionary computation. The fact that each client is autonomous leads to a complex behavior that will be examined in the work, so that the design can be validated, rules of thumb can be extracted, and the limits of scalability can be found. In this paper we advance the design of an asynchronous, fault-tolerant and scalable distributed evolutionary algorithm based on the object store CouchDB. We test experimentally the different options and show the trade-offs that pool and island-based solutions offer. © 2012 IEEE.


Mora A.M.,University of Granada | De Las Cuevas P.,University of Granada | Merelo J.J.,University of Granada | Zamarripa S.,S2 Grupo | And 5 more authors.
Proceedings of the ACM Symposium on Applied Computing | Year: 2014

This work presents the description of the architecture of a novel enterprise security system, still in development, which can prevent and deal with the security flaws derived from the users in a company. Thus, the Multiplatform Usable Endpoint Security system (MUSES) considers diverse factors such as the information distribution, the type of accesses, the context where the users are, the category of users, or the mix between personal and private data, among others. This system includes an event correlator and a risk and trust analysis engine to perform the decision process. MUSES follows a set of defined security rules, according to the enterprise security policies, but it is able to self-adapt the decisions and even create new security rules depending on the user behaviour, the specific device, and the situation or context. To this aim MUSES applies machine learning and computational intelligence techniques which can also be used to predict potential unsafe or dangerous user's behaviour. Copyright 2014 ACM.


Merelo-Guervos J.-J.,University of Granada | Mora A.,University of Granada | Cruz J.A.,University of Habana | Esparcia A.I.,S2 Grupo
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2012

This work presents the mapping of an evolutionary algorithm to the CouchDB object store. This mapping decouples the population from the evolutionary algorithm, and allows a distributed and asynchronous operation of clients written in different languages. In this paper we present initial tests which prove that the novel algorithm design still performs as an evolutionary algorithm and try to find out what are the main issues concerning it, what kind of speedups should we expect, and how all this affects the fundamentals of the evolutionary algorithm. © 2012 Springer-Verlag.


Pagano D.,TU Munich | Juan M.A.,S2 Grupo | Bagnato A.,TXT E solutions | Roehm T.,TU Munich | And 2 more authors.
Proceedings - International Conference on Software Engineering | Year: 2012

Software maintenance and support services are key factors to the customer perception of software product quality. The overall goal of FastFix is to provide developers with a real-time maintenance environment that increases efficiency and reduces costs, improving accuracy in identification of failure causes and facilitating their resolution. To achieve this goal, FastFix observes application execution and user interaction at runtime. We give an overview of the functionality of FastFix and present one of its main application scenarios. © 2012 IEEE.


Esparcia-Alcazar A.I.,S2 Grupo | Moravec J.,Charles University
Soft Computing | Year: 2013

Estimating the fitness value of individuals in an evolutionary algorithm in order to reduce the computational expense of actually calculating the fitness has been a classical pursuit of practitioners. One area which could benefit from progress in this endeavour is bot evolution, i.e. the evolution of non-playing characters in computer games. Because assigning a fitness value to a bot (or rather, the decision tree that controls its behaviour) requires playing the game, the process is very costly. In this work, we introduce two major contributions to speed up this process in the computer game Unreal Tournament 2004™. Firstly, a method for fitness value approximation in genetic programming which is based on the idea that individuals that behave in a similar fashion will have a similar fitness. Thus, similarity of individuals is taken at the performance level, in contrast to commonly employed approaches which are either based on similarity of genotypes or, less frequently, phenotypes. The approximation performs a weighted average of the fitness values of a number of individuals, attaching a confidence level which is based on similarity estimation. The latter is the second contribution of this work, namely a method for estimating the similarity between individuals. This involves carrying out a number of tests consisting of playing a 'static' version of the game (with fixed inputs) with the individuals whose similarity is under evaluation and comparing the results. Because the tests involve a limited version of the game, the computational expense of the similarity estimation plus that of the fitness approximation is much lower than that of directly calculating the fitness. The success of the fitness approximation by similarity estimation method for bot evolution in UT2K4 allows us to expect similar results in environments that share the same characteristics. © 2012 Springer-Verlag Berlin Heidelberg.


Alfaro-Cid E.,Polytechnic University of Valencia | Sharman K.,Polytechnic University of Valencia | Esparcia-Alcazar A.I.,S2 Grupo
Evolutionary Computation | Year: 2014

This work describes an approach devised by the authors for time series classification. In our approach genetic programming is used in combination with a serial processing of data, where the last output is the result of the classification. The use of genetic programming for classification, although still a field where more research in needed, is not new. However, the application of genetic programming to classification tasks is normally done by considering the input data as a feature vector. That is, to the best of our knowledge, there are not examples in the genetic programming literature of approaches where the time series data are processed serially and the last output is considered as the classification result. The serial processing approach presented here fills a gap in the existing literature. This approach was tested in three different problems. Two of them are real world problems whose data were gathered for online or conference competitions. As there are published results of these two problems this gives us the chance to compare the performance of our approach against top performing methods. The serial processing of data in combination with genetic programming obtained competitive results in both competitions, showing its potential for solving time series classification problems. The main advantage of our serial processing approach is that it can easily handle very large datasets. © 2014 by the Massachusetts Institute of Technology.


Barcelo-Rico F.,S2 Grupo | Esparcia-Alcazar A.I.,S2 Grupo | Villalon-Huerta A.,S2 Grupo | Villalon-Huerta A.,Polytechnic University of Valencia
Studies in Computational Intelligence | Year: 2015

Advanced Persistent Threats (APTs) are a highly sophisticated type of cyber attack usually aimed at large and powerful organisations.Human expert knowledge, coded as rules, can be used to detect these attacks when they attempt to extract information of their victim hidden within normal http traffic. Often, experts base their decisions on anomaly detection techniques, working under the hypothesis that APTs generate traffic that differs from normal traffic. In this work we aim at developing classifiers that can help human experts to findAPTs.We first define an anomaly score metric to select the most anomalous subset of traffic data; then the human expert labels the instances within this set; finally we train a classifier using both labeled and unlabelled data. Three computational intelligence methods were employed to train classifiers, namely genetic programming, decision trees and support vector machines. The results show their potential in the fight against APTs. © Springer International Publishing Switzerland 2016.

Loading S2 Grupo collaborators
Loading S2 Grupo collaborators