Buscador

Encontrado(s) 105833 resultado(s)
Encontrada(s) 10584 página(s)

Introducing a Robust and Efficient Stopping Criterion for MOEA's

  • Guerrero Madrid, José Luis
  • Martí, Luis
  • Berlanga de Jesús, Antonio
  • García, Jesús
  • Molina López, José Manuel
Proceedings of: IEEE World Congress on Computational Intelligence 2010 (WCCI 2010): IEEE Congress on Evolutionary Computation (CEC 2010). Barcelona, Spain, 18-23 July 2010., Soft computing methods, and Multi-Objective Evolutionary Algorithms (MOEAs) in particular, lack a general convergence criterion which prevents these algorithms from detecting the generation where further evolution will provide little improvements (or none at all) over the current solution, making them waste computational resources. This paper presents the Least Squares Stopping Criterion (LSSC), an easily configurable and implementable, robust and efficient stopping criterion, based on simple statistical parameters and residue analysis, which tries to introduce as few setup parameters as possible, being them always related to the MOEAs research field rather than the techniques applied by the criterion., This work was supported in part by Projects CICYT TIN2008-06742-C02-02/TSI, CICYT TEC2008-06732-C02-02/TEC, CAM CONTEXTS (S2009/TIC- 1485) and DPS2008-07029-C02-02, Publicado
Proyecto:


Amar en tiempos revueltos: métodos de trabajo y razones de un éxito

  • Molina Cañabate, Juan Pedro
  • Lara Padilla, María Tíscar
  • Lamuedra Graván, María
Amar en tiempos revueltos, serie producida por Diagonal TV y que emite Televisión Española desde hace cuatro años, es una de las series más vistas de la cadena estatal española, llegando a unos shares del 29,9 por ciento. Ente otras distinciones, obtuvo la Medalla de Plata a la mejor serie dramática en el Festival Internacional de Programas de Televisión y Promociones de Nueva York en 2007 y fue finalista en el TVFest de Monte Carlo en 2008. Próximamente, Amar en tiempos revueltos podrá verse en su versión original en Estados Unidos y América Latina gracias al acuerdo alcanzado entre Diagonal TV y Telemundo (propiedad de NBC Universal). En apariencia, el éxito de esta serie radica en causas perceptibles, manifiestas, como el trabajo artístico de los actores y su dirección, la escenografía, el vestuario, o unos guiones bien tramados y con buenos diálogos. Sin embargo, para que todos estos elementos puedan unirse para crear un producto de éxito, es necesaria una gestión que va más allá de lo empresarial y que está basada en la comunicación interna., Amar en tiempos revueltos, produced by Diagonal TV and shown by the Spanish Public Service Broadcasting Corporation, is one of Televisión Española's most popular series with a 20.8 per cent share. Its success can be put down to the skill a nd expertise of both actors and directors and the care given to both sets and wardrobe, together with the well constructed plots and the quality of the dialogue. However, a successful product of this kind requires a management style which goes beyond straightforward business considerations and is also based on excellent in-house communication., Publicado
Proyecto:


A Multi-Agent Approach for Provisioning of e-Services in u-Commerce Environments

  • Sánchez-Pi, Nayat
  • Molina, José M.
Purpose: Taking into account the importance of e-commerce and the current applications of AI techniques in this area, this research aims to adequate the design of a multi-agent system for the provisioning of e-services in u-commerce environments. This proposal is centred on the methods of evaluation in a u-e-commerce environment. Design/methodology/approach: The multi-agent systems (MAS) approach is based on an MAS model developed for AmI that has been redesigned to support u-commerce. The use of a recommendation system, previously developed by the research group, is suggested for this MAS. The methodological proposal centres on the evaluation of this type of system. Findings: The evaluation of this type of system is the principal problem of current research. Therefore, this is the main contribution of the paper. Research limitations/implications: The different evaluation methods that are proposed, whether qualitative or quantitative, offer the possibility of measuring the added value that the context can give to the use of e-services in different domains of application. Qualitative evaluation should consider the customer as a central piece in the system. In addition, quantitative methods should objectively evaluate the contribution of context to the application. Practical implications: At present, there is no single method for evaluating the benefits of different u-commerce systems, so a new method needs to be found based on these techniques. Originality/value: The research proposes an MAS designed for u-commerce domains, analyzes the capacity of trust management techniques in this environment, and proposes several evaluation methods to show the benefits of context information in the use of e-services. Several real developments are described to show the different applications of MAS in u-commerce and how evaluation is carried out., This work has been partially supported by Projects CICYT TIN2008-06742-C02-02/TSI, CICYT TEC2008-06732-C02-02/TEC, SINPROB, CAM CONTEXT and DPS2008-07029-C02-02., Publicado
Proyecto:


Finding an Evolutionarily Stable Strategy in Agent Reputation and Trust (ART) 2007 Competition

  • Carbó Rubiera, Javier Ignacio
  • Molina López, José Manuel
Proceedings of: 23rd International Conference on Industrial Engineering and Other Applications of Applied intelligent Systems, IEA/AIE 2010, Cordoba, Spain, June 1-4, 2010, Our proposal is to apply a Game Theoretic approach to the games played in Agent Reputation and Trust Final Competitions. Using such testbed, three international competitions were successfully carried out jointly with the last AAMAS international Conferences. The corresponding way to define the winner of such competitions was to run a game with all the participants (16). Our point is that such game does not represent a complete way to determine the best trust/reputation strategy, since it is not proved that such strategy is evolutionarily stable. Specifically we prove that when the strategy of the winner of the two first international competitions (2006 and 2007) becomes dominant, it is defeated by other participant trust strategies. Then we found out (through a repeated game definition) the right equilibrium of trust strategies that is evolutionarily stable. This kind of repeated game has to be taken into account in the evaluation of trust strategies, and this conclusion would improve the way trust strategies have to be compared., This work was supported in part by Projects CICYT TIN2008-06742-C02-02/ TSI, CICYT TEC2008-06732-C02-02/TEC, SINPROB, CAM MADRINET S-0505/TIC/0255 and DPS2008-07029-C02-02., Publicado
Proyecto:


Analysis of a sensor fusion hybrid solution for indoor/outdoor robot navigation

  • Martí, Enrique David
  • García, Jesús
  • Molina, José M.
Proceedings of: 5th ESA Workshop on Satellite Navigation Technologies and European Workshop on GNSS Signals and Signal Processing (NAVITEC 2010). Noordwij, Netherlands, 8-10 December 2010, Autonomous mobile robots need robust, flexible and accurate navigation algorithms. One approach consists in fusing as many information sources as possible, integrating measures from internal sensors with data obtained from external sensing entities. This work presents a solution for combined indoor/outdoor robot navigation, and analyzes some preliminary results in an outdoor environment using a Particle Filter for GPS/INS sensor fusion. Experiments are based in predesigned trajectories which have been simulated in first place and then reproduced using a robotic platform. As a concluding remark, some considerations about the use of Particle Filters and the differences between simulated and real data are presented, This work was supported in part by Projects ATLANTIDA, CICYT TIN2008-06742-C02-02/TSI, CICYT TEC2008- 06732-C02-02/TEC, SINPROB, CAM MADRINET S- 0505/TIC/0255 DPS2008-07029-C02-02., Publicado
Proyecto:


Air traffic trajectories segmentation based on time-series sensor data

  • Guerrero Madrid, José Luis
  • García, Jesús
  • Molina, José M.
Proyecto:


Sentando las bases para la configuración de un modelo viable de juez

  • Aguinaga Meza, Ernesto Alonso
Máster Oficial en Estudios Avanzados en Derechos Humanos, Presidente: Javier Ansuátegui; Vocal: Emilio Pajares; Secretaria: Silvina Ribotta
Proyecto:


Discurso del Rector [Acto Día de la Universidad 2009/2010]

  • Peña, Daniel
Discurso del Rector en el Día de la Universidad celebrado el 28 de enero de 2010
Proyecto:


Robotic Motion using Harmonic Functions and Finite Elements

  • Garrido, Santiago
  • Moreno Lorente, Luis Enrique
  • Blanco Rojas, María Dolores
  • Martín Monar, Fernando
The harmonic functions have proved to be a powerful technique for motion planning in a known environment. They have two important properties: given an initial point and an objective in a connected domain, a unique path exists between those points. This path is the maximum gradient path of the harmonic function that begins in the initial point and ends in the goal point. The second property is that the harmonic function cannot have local minima in the interior of the domain (the objective point is considered as a border). This paper proposes a new method to solve Laplace's equation. The harmonic function solution with mixed boundary conditions provides paths that verify the smoothness and safety considerations required for mobile robot path planning. The proposed approach uses the Finite Elements Method to solve Laplace's equation, and this allows us to deal with complicated shapes of obstacles and walls. Mixed boundary conditions are applied to the harmonic function to improve the quality of the trajectories. In this way, the trajectories are smooth, avoiding the corners of walls and obstacles, and the potential slope is not too small, avoiding the difficulty of the numerical calculus of the trajectory. Results show that this method is able to deal with moving obstacles, and even for non-holonomic vehicles. The proposed method can be generalized to 3D or more dimensions and it can be used to move robot manipulators.
Proyecto:


Moving away from error-based learning in multi-objective estimation of distribution algorithms

  • Martí, Luis
  • García, Jesús
  • Berlanga de Jesús, Antonio
  • Molina López, José Manuel
Proceedings of: 12th annual conference on Genetic and evolutionary computation (GECCO '10). Portland, Oregon, USA, July 7-11, 2010., In this work we analyze the model-building issue and the requirements it imposes on the learning paradigm being used. We argue that error-based learning, the class of learning most commonly used in MOEDAs, is responsible for current MOEDA underachievement. We present ART as a viable alternative and present a novel algorithm called multi-objective ART-based EDA (MARTEDA) that uses a Gaussian ART neural network for model-building and an hypervolume based selector as described for the HypE algorithm. We experimentally show that thanks to MARTEDA's novel model-building approach and an indicator-based population ranking the algorithm it is able to outperform similar MOEDAs and MOEAs., This work was supported in part by projects CICYT TIN2008-06742-C02-02/TSI, CICYT TEC2008-06732-C02-02/TEC, DPS2008-07029-C02-02 and CAM CONTEXTS S2009/TIC-1485., Publicado
Proyecto:


Buscador avanzado