Oscar Cordon - Profile on Academia.edu (original) (raw)
Papers by Oscar Cordon
arXiv (Cornell University), Nov 24, 2015
An appropriate visualization of multiobjective nondominated solutions is a valuable asset for dec... more An appropriate visualization of multiobjective nondominated solutions is a valuable asset for decision making. Although there are methods for visualizing the solutions in the design space, they do not provide any information about their relationship. In this work, we propose a novel methodology that allows the visualization of the non-dominated solutions in the design space and their relationships by means of a network. The nodes represent the solutions in the objective space, while the edges show the relationships between the solutions in the design space. Our proposal (called moGrams) thus provides a joint visualization of both objective and design spaces. It aims at helping the decision maker to get more understanding of the problem so that (s)he can choose the more appropriate final solution. moGrams can be applied to any multicriteria problem in which the solutions are related by a similarity metric. Besides, the decision maker interaction is facilitated by modifying the network based on the current preferences to obtain a clearer view. An exhaustive experimental study is performed using three multiobjective problems in order to show both the usefulness and versatility of moGrams. The results exhibit interesting characteristics of our methodology for visualizing and analyzing solutions of multiobjective problems.
Cuadernos De Medicina Forense, Oct 1, 2008
Se presenta un nuevo método para la identificación humana por medio de análisis de imágenes y sup... more Se presenta un nuevo método para la identificación humana por medio de análisis de imágenes y superposición fotográfica. Se discuten las ventajas que aporta la incorporación de imágenes tridimensionales del cráneo, ya que, por un lado, facilitan la superposición y por otro, eliminan errores de escalado al estar el modelo 3D a tamaño real. Este trabajo se ha desarrollado por un equipo interdisciplinar y su objetivo fundamental es proporcionar una herramienta semiautomática de identificación humana, basada en el reconocimiento craneofacial. Palabras clave: Identificación humana, Superposición de imágenes, Análisis tridimensional.
Ant Colony Optimisation: models and applications
Soft Computing, 2002
Abstract top Ant Colony Optimization (ACO) is a metaheuristic that is inspired by the shortest pa... more Abstract top Ant Colony Optimization (ACO) is a metaheuristic that is inspired by the shortest path searching behavior of various ant species [1, 2]. The initial work of Dorigo, Maniezzo and Colorni [3, 4] who proposed the first ACO algorithm called Ant System, has stimulated a still strongly increasing number of researchers to develop more sophisticated and better performing ACO algorithms that are used to successfully solve a large number of hard combinatorial optimization problems such as the traveling salesman problem, the ...
Genetic Fuzzy Rule-Based Systems Based on the Iterative Rule Learning Approach
WORLD SCIENTIFIC eBooks, Jul 1, 2001
Some Real-world Applications of Soft Artificial Intelligence: Scientogram Mining, Assembly Line Balancing, and Forensic Identification
CCIA, 2012
Soft artificial intelligence methods keep on tackling classical artificial intelligence problems su... more Soft artificial intelligence methods keep on tackling classical artificial intelligence problems such as heuristic search, knowledge representation, and machine learning. In this lecture we will review three different real-world applications solved by means of modern heuristic search techniques (metaheuristics) as well as by the use of fuzzy sets to represent the inherent imprecission of the available knowledge. First, the analysis and comparison of visual science maps (scientograms) will be tackled as a single-and multiobjective graph ...
On the impact of Distance-based Relative Competence Weighting approach in One-vs-One classification for Evolutionary Fuzzy Systems: DRCW-FH-GBML algorithm
The advantages of multi-classification schemes based on decomposition strategies, and especially ... more The advantages of multi-classification schemes based on decomposition strategies, and especially the One-vs-One framework, have been stressed even for those algorithms that can address multiple classes. However, there is an inherent hitch for the One-vs-One learning scheme related to the decision process: the non-competent classifier problem. This issue refers to the case where a binary classifier outputs a score degree for a couple of classes that are not related with the input example, thus including “noise” in the score-matrix and degrading the final accuracy. For this reason, several approaches have been developed in order to address the influence of the non-competence. Among them, the distance-based combination strategy has excelled as a very robust solution. In this contribution, we aim at investigating the behaviour of this approach using Evolutionary Fuzzy Systems as baseline classifiers. We will show that the synergy between both methodologies allows a significant improvement of the results to be obtained in contrast to the standard classifier and the classical One-vs-One scheme.
Guest editorial "Ant colony optimization : models and applications
Soft Computing, 2002
info:eu-repo/semantics/publishe
European Society for Fuzzy Logic and Technology Conference, 2001
This paper presents an evolutionary learning process for linguistic modeling with weighted double... more This paper presents an evolutionary learning process for linguistic modeling with weighted double-consequent fuzzy rules. These kinds of fuzzy rules are used to improve the linguistic modeling, with the aim of introducing a trade-off between interpretability and precision. The use of weighted double-consequent fuzzy rules makes more complex the modeling and learning process, increasing the solution search space. Therefore, the cooperative coevolution, an advanced evolutionary technique proposed to solve decomposable complex problems, is considered to learn these kinds of rules. The proposal has been tested with different problems achieving good results.
Special issue on Ant Colony Optimization
Soft Computing, 2002
DI-fusion, le Dépôt institutionnel numérique de l'ULB, est l&... more DI-fusion, le Dépôt institutionnel numérique de l'ULB, est l'outil de référencementde la production scientifique de l'ULB.L'interface de recherche DI-fusion permet de consulter les publications des chercheurs de l'ULB et les thèses qui y ont été défendues.
IEEE Computational Intelligence Magazine, May 1, 2016
Incorporating awareness and genetic-based viral marketing strategies to a consumer behavior model
In this paper, we will use agent-based modeling with the aim of simulating customer purchase proc... more In this paper, we will use agent-based modeling with the aim of simulating customer purchase processes in competitive market environments. These simulations will help to understand how customer behavior is affected by different social network topologies and the acquisition success of the products offered. We will extend a well-known agent-based model by incorporating an awareness customer behavior into it to make it closer to reality. In this way, individuals will not initially have a complete knowledge about all the products but they will gradually acquire it through a word-of-mouth process within the social network. Additionally, we will use genetic algorithms to generate automatic viral marketing strategies based on social network analysis metrics. We will compare the marketing results of the combined strategies and their impact based on different types of networks and the number of influential individuals. Finally, we will show that word-of-mouth evolves slower due to the awareness filter and that the genetic algorithm is able to find good solutions for targeting the most influential members of the market according to social network information.
A framework of opinion dynamics using fuzzy linguistic 2-tuples
Knowledge Based Systems, Dec 1, 2021
IEEE Symposium Series on Computational Intelligence, Dec 1, 2015
International Journal of Intelligent Systems, 2020
Understanding consumer behaviors and how consumers react to marketing campaigns and viral word-of... more Understanding consumer behaviors and how consumers react to marketing campaigns and viral word-of-mouth processes is crucial for marketers. Classical approaches try to infer this information from a global top-down perspective. However, a more suitable and natural approach is to model consumer behaviors in a heterogeneous and decentralized bottom-up approach. In this case, each virtual consumer has her own mental state and decision-making strategies to simulate her purchase decisions. The system of virtual consumers generates the global sales and a marketer can understand the rules that govern the market. A well-known paradigm to model these systems is agent-based modeling (ABM). In this manuscript we present an ABM where the brand preferences of the consumer agents are modeled using 2-tuple fuzzy linguistic variables. These variables represent the perceptions these consumers have on the different aspects or drivers every product available in the market has (e.g., price or quality). The product selection process of the agents is based on those perceptions and a utility maximization rule. This rule requires a fuzzy aggregation of the fuzzy linguistic perceptions about the products. Our proposal employs an ordered weighted average (OWA) to aggregate them. Our experiments show this approach does not suffer any loss of information when applied on data from real markets. Hence it is a suitable representation of the products preferences, normally represented by qualitative values in marketing surveys. To the best of our knowledge, this is the first work integrating a marketing ABM with fuzzy linguistic modeling.
rESumEn. las universidades tienen una responsabilidad formativa para con su estudiantado pero tam... more rESumEn. las universidades tienen una responsabilidad formativa para con su estudiantado pero también tienen una responsabilidad de retorno del conocimiento generado en ellas a la sociedad. Habitualmente, ese retorno de conocimiento se lleva a cabo mediante muestras, exposiciones y conferencias abiertas al público en general. Sin embargo, todas estas actividades de divulgación están limitadas geográficamente. tanto para la formación reglada como para el aprendizaje a lo largo de toda la vida (lifelong learning), la metodología mooC ha demostrado ser una forma eficaz para cumplir con esta responsabilidad, tanto en entornos estructurados como en el aprendizaje continuo. En cualquiera de los casos, el despliegue de open Elearning resources (oEr) y de massive online open Courses (mooCs) por parte de una institución educativa conlleva una responsabilidad para la institución en sí, lo que hace necesario establecer ciertos controles sobre la calidad del producto generado, sobre todo en lo referente a su estructura y al desempeño de las personas que transmiten el conocimiento y que apoyan a dicha transmisión. En este artículo se repasan todos los aspectos referentes a los procedimientos seguidos para la creación de mooCs en una experiencia particular: la Iniciativa mooCs ugr. aBStraCt. universities have a formative responsibility towards their students but they also have a responsibility to return the knowledge generated in them to society. usually, this return of knowledge is carried out through exhibitions, exhibitions and conferences open to the general public. However, all of these outreach activities are geographically limited. for both regulated training and lifelong learning (lifelong learning), the mooC methodology has proven to be an effective way to fulfill this responsibility, both in structured environments and in continuous learning. In any case, the deployment of open Elearning resources (oEr) and massive online open Courses (mooCs) by an educational institution entails a responsibility for the institution itself, which makes it necessary to establish certain controls over the quality of the generated product, especially in relation to its structure and the performance of the people who transmit the knowledge and who support such transmission. In this article we review all the aspects referring to the procedures followed for the creation of mooCs in a particular experience: the mooCs ugr Initiative.
arXiv (Cornell University), Sep 13, 2021
Human sciences have traditionally relied on human reasoning and intelligence to infer knowledge f... more Human sciences have traditionally relied on human reasoning and intelligence to infer knowledge from a wide range of sources, such as oral and written narrations, reports, and traditions. Here we develop an extension of classical social network analysis approaches to incorporate the concept of meaning in each actor, as a mean to quantify and infer further knowledge from the original source of the network. This extension is based on a new affinity function, the semantic affinity, that establishes fuzzy-like relationships between the different actors in the network, using combinations of affinity functions. We also propose a new heuristic algorithm based on the shortest capacity problem to compute this affinity function. We use these concept of meaning and semantic affinity to analyze and compare the gods and heroes from three different classical mythologies: Greek, Celtic and Nordic. We study the relationships of each individual mythology and those of common structure that is formed when we fuse the three of them. We show a strong connection between the Celtic and Nordic gods and that Greeks put more emphasis on heroic characters rather than deities. Our approach provides a technique to highlight and quantify important relationships in the original domain of the network not deducible from its structural properties.
12th On-line World Conference on Soft Computing in Industrial Applications
IEEE Computational Intelligence Magazine, Feb 1, 2019
Evolutionary fuzzy systems are one of the greatest advances within the area of computational inte... more Evolutionary fuzzy systems are one of the greatest advances within the area of computational intelligence. They consist of evolutionary algorithms applied to the design of fuzzy systems. Thanks to this hybridization, superb abilities are provided to fuzzy modeling in many different data science scenarios. This contribution is intended to comprise a position paper developing a comprehensive analysis of the evolutionary fuzzy systems research field. To this end, the "4 W" questions are posed and addressed with the aim of understanding the current context of this topic and its significance. Specifically, it will be pointed out why evolutionary fuzzy systems are important from an explainable point of view, when they began, what they are used for, and where the attention of researchers should be directed to in the near future in this area. They must play an important role for the emerging area of eXplainable Artificial Intelligence (XAI) learning from data.
Lecture Notes in Computer Science, 2022
In this work, we propose a novel architecture for face age editing that can produce structural mo... more In this work, we propose a novel architecture for face age editing that can produce structural modifications while maintaining relevant details present in the original image. We disentangle the style and content of the input image and propose a new decoder network that adopts a style-based strategy to combine the style and content representations of the input image while conditioning the output on the target age. We go beyond existing aging methods allowing users to adjust the degree of structure preservation in the input image during inference. To this purpose, we introduce a masking mechanism, the CUstom Structure Preservation module, that distinguishes relevant regions in the input image from those that should be discarded. CUSP requires no additional supervision. Finally, our quantitative and qualitative analysis which include a user study, show that our method outperforms prior art and demonstrates the effectiveness of our strategy regarding image editing and adjustable structure preservation. Code and pretrained models are available at https://github.com/guillermogotre/CUSP.
Information Sciences
In the bounded confidence framework, agents' opinions evolve as a result of interactions with oth... more In the bounded confidence framework, agents' opinions evolve as a result of interactions with other agents having similar opinions. Thus, consensus or fragmentation of opinions can be reached, but not extremization (the evolution of opinions towards an extreme value). In contrast, when repulsion mechanisms are at work, agents with distant opinions interact and repel each other, leading to extremization. This work proposes a general opinion dynamics framework of bounded confidence and repulsion, which includes social network interactions and agent-independent time-varying rationality. We extensively analyze the performance of our model to show that the degree of extremization among a population can be controlled by the repulsion rule, and social networks promote extreme opinions. Agent-based rationality and time-varying adaptation also bear a strong impact on opinion dynamics. The high accuracy of our model is determined in a real-world social network well referenced in the literature, the Zachary Karate Club (with a known ground truth). Finally, we use our model to analyze the extremization of opinions in a real-world scenario, in Spain: a marketing action for the Netflix series ''Narcos".
arXiv (Cornell University), Nov 24, 2015
An appropriate visualization of multiobjective nondominated solutions is a valuable asset for dec... more An appropriate visualization of multiobjective nondominated solutions is a valuable asset for decision making. Although there are methods for visualizing the solutions in the design space, they do not provide any information about their relationship. In this work, we propose a novel methodology that allows the visualization of the non-dominated solutions in the design space and their relationships by means of a network. The nodes represent the solutions in the objective space, while the edges show the relationships between the solutions in the design space. Our proposal (called moGrams) thus provides a joint visualization of both objective and design spaces. It aims at helping the decision maker to get more understanding of the problem so that (s)he can choose the more appropriate final solution. moGrams can be applied to any multicriteria problem in which the solutions are related by a similarity metric. Besides, the decision maker interaction is facilitated by modifying the network based on the current preferences to obtain a clearer view. An exhaustive experimental study is performed using three multiobjective problems in order to show both the usefulness and versatility of moGrams. The results exhibit interesting characteristics of our methodology for visualizing and analyzing solutions of multiobjective problems.
Cuadernos De Medicina Forense, Oct 1, 2008
Se presenta un nuevo método para la identificación humana por medio de análisis de imágenes y sup... more Se presenta un nuevo método para la identificación humana por medio de análisis de imágenes y superposición fotográfica. Se discuten las ventajas que aporta la incorporación de imágenes tridimensionales del cráneo, ya que, por un lado, facilitan la superposición y por otro, eliminan errores de escalado al estar el modelo 3D a tamaño real. Este trabajo se ha desarrollado por un equipo interdisciplinar y su objetivo fundamental es proporcionar una herramienta semiautomática de identificación humana, basada en el reconocimiento craneofacial. Palabras clave: Identificación humana, Superposición de imágenes, Análisis tridimensional.
Ant Colony Optimisation: models and applications
Soft Computing, 2002
Abstract top Ant Colony Optimization (ACO) is a metaheuristic that is inspired by the shortest pa... more Abstract top Ant Colony Optimization (ACO) is a metaheuristic that is inspired by the shortest path searching behavior of various ant species [1, 2]. The initial work of Dorigo, Maniezzo and Colorni [3, 4] who proposed the first ACO algorithm called Ant System, has stimulated a still strongly increasing number of researchers to develop more sophisticated and better performing ACO algorithms that are used to successfully solve a large number of hard combinatorial optimization problems such as the traveling salesman problem, the ...
Genetic Fuzzy Rule-Based Systems Based on the Iterative Rule Learning Approach
WORLD SCIENTIFIC eBooks, Jul 1, 2001
Some Real-world Applications of Soft Artificial Intelligence: Scientogram Mining, Assembly Line Balancing, and Forensic Identification
CCIA, 2012
Soft artificial intelligence methods keep on tackling classical artificial intelligence problems su... more Soft artificial intelligence methods keep on tackling classical artificial intelligence problems such as heuristic search, knowledge representation, and machine learning. In this lecture we will review three different real-world applications solved by means of modern heuristic search techniques (metaheuristics) as well as by the use of fuzzy sets to represent the inherent imprecission of the available knowledge. First, the analysis and comparison of visual science maps (scientograms) will be tackled as a single-and multiobjective graph ...
On the impact of Distance-based Relative Competence Weighting approach in One-vs-One classification for Evolutionary Fuzzy Systems: DRCW-FH-GBML algorithm
The advantages of multi-classification schemes based on decomposition strategies, and especially ... more The advantages of multi-classification schemes based on decomposition strategies, and especially the One-vs-One framework, have been stressed even for those algorithms that can address multiple classes. However, there is an inherent hitch for the One-vs-One learning scheme related to the decision process: the non-competent classifier problem. This issue refers to the case where a binary classifier outputs a score degree for a couple of classes that are not related with the input example, thus including “noise” in the score-matrix and degrading the final accuracy. For this reason, several approaches have been developed in order to address the influence of the non-competence. Among them, the distance-based combination strategy has excelled as a very robust solution. In this contribution, we aim at investigating the behaviour of this approach using Evolutionary Fuzzy Systems as baseline classifiers. We will show that the synergy between both methodologies allows a significant improvement of the results to be obtained in contrast to the standard classifier and the classical One-vs-One scheme.
Guest editorial "Ant colony optimization : models and applications
Soft Computing, 2002
info:eu-repo/semantics/publishe
European Society for Fuzzy Logic and Technology Conference, 2001
This paper presents an evolutionary learning process for linguistic modeling with weighted double... more This paper presents an evolutionary learning process for linguistic modeling with weighted double-consequent fuzzy rules. These kinds of fuzzy rules are used to improve the linguistic modeling, with the aim of introducing a trade-off between interpretability and precision. The use of weighted double-consequent fuzzy rules makes more complex the modeling and learning process, increasing the solution search space. Therefore, the cooperative coevolution, an advanced evolutionary technique proposed to solve decomposable complex problems, is considered to learn these kinds of rules. The proposal has been tested with different problems achieving good results.
Special issue on Ant Colony Optimization
Soft Computing, 2002
DI-fusion, le Dépôt institutionnel numérique de l'ULB, est l&... more DI-fusion, le Dépôt institutionnel numérique de l'ULB, est l'outil de référencementde la production scientifique de l'ULB.L'interface de recherche DI-fusion permet de consulter les publications des chercheurs de l'ULB et les thèses qui y ont été défendues.
IEEE Computational Intelligence Magazine, May 1, 2016
Incorporating awareness and genetic-based viral marketing strategies to a consumer behavior model
In this paper, we will use agent-based modeling with the aim of simulating customer purchase proc... more In this paper, we will use agent-based modeling with the aim of simulating customer purchase processes in competitive market environments. These simulations will help to understand how customer behavior is affected by different social network topologies and the acquisition success of the products offered. We will extend a well-known agent-based model by incorporating an awareness customer behavior into it to make it closer to reality. In this way, individuals will not initially have a complete knowledge about all the products but they will gradually acquire it through a word-of-mouth process within the social network. Additionally, we will use genetic algorithms to generate automatic viral marketing strategies based on social network analysis metrics. We will compare the marketing results of the combined strategies and their impact based on different types of networks and the number of influential individuals. Finally, we will show that word-of-mouth evolves slower due to the awareness filter and that the genetic algorithm is able to find good solutions for targeting the most influential members of the market according to social network information.
A framework of opinion dynamics using fuzzy linguistic 2-tuples
Knowledge Based Systems, Dec 1, 2021
IEEE Symposium Series on Computational Intelligence, Dec 1, 2015
International Journal of Intelligent Systems, 2020
Understanding consumer behaviors and how consumers react to marketing campaigns and viral word-of... more Understanding consumer behaviors and how consumers react to marketing campaigns and viral word-of-mouth processes is crucial for marketers. Classical approaches try to infer this information from a global top-down perspective. However, a more suitable and natural approach is to model consumer behaviors in a heterogeneous and decentralized bottom-up approach. In this case, each virtual consumer has her own mental state and decision-making strategies to simulate her purchase decisions. The system of virtual consumers generates the global sales and a marketer can understand the rules that govern the market. A well-known paradigm to model these systems is agent-based modeling (ABM). In this manuscript we present an ABM where the brand preferences of the consumer agents are modeled using 2-tuple fuzzy linguistic variables. These variables represent the perceptions these consumers have on the different aspects or drivers every product available in the market has (e.g., price or quality). The product selection process of the agents is based on those perceptions and a utility maximization rule. This rule requires a fuzzy aggregation of the fuzzy linguistic perceptions about the products. Our proposal employs an ordered weighted average (OWA) to aggregate them. Our experiments show this approach does not suffer any loss of information when applied on data from real markets. Hence it is a suitable representation of the products preferences, normally represented by qualitative values in marketing surveys. To the best of our knowledge, this is the first work integrating a marketing ABM with fuzzy linguistic modeling.
rESumEn. las universidades tienen una responsabilidad formativa para con su estudiantado pero tam... more rESumEn. las universidades tienen una responsabilidad formativa para con su estudiantado pero también tienen una responsabilidad de retorno del conocimiento generado en ellas a la sociedad. Habitualmente, ese retorno de conocimiento se lleva a cabo mediante muestras, exposiciones y conferencias abiertas al público en general. Sin embargo, todas estas actividades de divulgación están limitadas geográficamente. tanto para la formación reglada como para el aprendizaje a lo largo de toda la vida (lifelong learning), la metodología mooC ha demostrado ser una forma eficaz para cumplir con esta responsabilidad, tanto en entornos estructurados como en el aprendizaje continuo. En cualquiera de los casos, el despliegue de open Elearning resources (oEr) y de massive online open Courses (mooCs) por parte de una institución educativa conlleva una responsabilidad para la institución en sí, lo que hace necesario establecer ciertos controles sobre la calidad del producto generado, sobre todo en lo referente a su estructura y al desempeño de las personas que transmiten el conocimiento y que apoyan a dicha transmisión. En este artículo se repasan todos los aspectos referentes a los procedimientos seguidos para la creación de mooCs en una experiencia particular: la Iniciativa mooCs ugr. aBStraCt. universities have a formative responsibility towards their students but they also have a responsibility to return the knowledge generated in them to society. usually, this return of knowledge is carried out through exhibitions, exhibitions and conferences open to the general public. However, all of these outreach activities are geographically limited. for both regulated training and lifelong learning (lifelong learning), the mooC methodology has proven to be an effective way to fulfill this responsibility, both in structured environments and in continuous learning. In any case, the deployment of open Elearning resources (oEr) and massive online open Courses (mooCs) by an educational institution entails a responsibility for the institution itself, which makes it necessary to establish certain controls over the quality of the generated product, especially in relation to its structure and the performance of the people who transmit the knowledge and who support such transmission. In this article we review all the aspects referring to the procedures followed for the creation of mooCs in a particular experience: the mooCs ugr Initiative.
arXiv (Cornell University), Sep 13, 2021
Human sciences have traditionally relied on human reasoning and intelligence to infer knowledge f... more Human sciences have traditionally relied on human reasoning and intelligence to infer knowledge from a wide range of sources, such as oral and written narrations, reports, and traditions. Here we develop an extension of classical social network analysis approaches to incorporate the concept of meaning in each actor, as a mean to quantify and infer further knowledge from the original source of the network. This extension is based on a new affinity function, the semantic affinity, that establishes fuzzy-like relationships between the different actors in the network, using combinations of affinity functions. We also propose a new heuristic algorithm based on the shortest capacity problem to compute this affinity function. We use these concept of meaning and semantic affinity to analyze and compare the gods and heroes from three different classical mythologies: Greek, Celtic and Nordic. We study the relationships of each individual mythology and those of common structure that is formed when we fuse the three of them. We show a strong connection between the Celtic and Nordic gods and that Greeks put more emphasis on heroic characters rather than deities. Our approach provides a technique to highlight and quantify important relationships in the original domain of the network not deducible from its structural properties.
12th On-line World Conference on Soft Computing in Industrial Applications
IEEE Computational Intelligence Magazine, Feb 1, 2019
Evolutionary fuzzy systems are one of the greatest advances within the area of computational inte... more Evolutionary fuzzy systems are one of the greatest advances within the area of computational intelligence. They consist of evolutionary algorithms applied to the design of fuzzy systems. Thanks to this hybridization, superb abilities are provided to fuzzy modeling in many different data science scenarios. This contribution is intended to comprise a position paper developing a comprehensive analysis of the evolutionary fuzzy systems research field. To this end, the "4 W" questions are posed and addressed with the aim of understanding the current context of this topic and its significance. Specifically, it will be pointed out why evolutionary fuzzy systems are important from an explainable point of view, when they began, what they are used for, and where the attention of researchers should be directed to in the near future in this area. They must play an important role for the emerging area of eXplainable Artificial Intelligence (XAI) learning from data.
Lecture Notes in Computer Science, 2022
In this work, we propose a novel architecture for face age editing that can produce structural mo... more In this work, we propose a novel architecture for face age editing that can produce structural modifications while maintaining relevant details present in the original image. We disentangle the style and content of the input image and propose a new decoder network that adopts a style-based strategy to combine the style and content representations of the input image while conditioning the output on the target age. We go beyond existing aging methods allowing users to adjust the degree of structure preservation in the input image during inference. To this purpose, we introduce a masking mechanism, the CUstom Structure Preservation module, that distinguishes relevant regions in the input image from those that should be discarded. CUSP requires no additional supervision. Finally, our quantitative and qualitative analysis which include a user study, show that our method outperforms prior art and demonstrates the effectiveness of our strategy regarding image editing and adjustable structure preservation. Code and pretrained models are available at https://github.com/guillermogotre/CUSP.
Information Sciences
In the bounded confidence framework, agents' opinions evolve as a result of interactions with oth... more In the bounded confidence framework, agents' opinions evolve as a result of interactions with other agents having similar opinions. Thus, consensus or fragmentation of opinions can be reached, but not extremization (the evolution of opinions towards an extreme value). In contrast, when repulsion mechanisms are at work, agents with distant opinions interact and repel each other, leading to extremization. This work proposes a general opinion dynamics framework of bounded confidence and repulsion, which includes social network interactions and agent-independent time-varying rationality. We extensively analyze the performance of our model to show that the degree of extremization among a population can be controlled by the repulsion rule, and social networks promote extreme opinions. Agent-based rationality and time-varying adaptation also bear a strong impact on opinion dynamics. The high accuracy of our model is determined in a real-world social network well referenced in the literature, the Zachary Karate Club (with a known ground truth). Finally, we use our model to analyze the extremization of opinions in a real-world scenario, in Spain: a marketing action for the Netflix series ''Narcos".