En la Sesión sobre “Videojuegos e Inteligencia Computacional”, celebrada recientemente en MAEB 2010, Valencia, presentamos la versión de lanzamiento del Videojuego Chapas, que utiliza la tecnología Genetic Terrain Programming para la generación de terrenos.
El proyecto en software libre tiene su forja accesible en: https://forja.unex.es/projects/chapas
Incluímos un vídeo del juego. Disfrútalo.
Here you are the AMAZING (:D) poster presented at EVO* 2010, inside the Workshop EVOGames.
The work describes the analysis, design and implementation of two evolutionary algorithms (a Genetic Algorithm and a Genetic Programming one) to improve the default Artificial Intelligence of the Bots (authonomous enemies) inside the PC Game Unreal.
It had plenty of visits and millards of interest people ask me about the topics in it…
… well, to be honest just, 10 people ask me (3 of them are friends of mine), but in some moments, around 8 or 10 people were looking at the poster and making jokes with me, since it seems to be a perfect mixture between a fabulous research study (:D) and a funny PC game.
Our paper “Multiobjective Genetic Programming Approach for a Smooth Modeling of the Release Kinetics of a Pheromone Dispenser” was presented at the workshop on Symbolic Regression and Modeling, part of the Genetic and Evolutionary Computation Conference, GECCO, held in Montreal, Canada, from July 8th ot 12th, 2009.
Here’s the abstract:
The accurate modeling of the release kinetics of pheromone dispensers is a matter or great importance for ensuring that the dispenser field-life covers the right period of the pest and for optimizing the layout of dispensers in the treated area.
A new experimental dispenser has been recently designed by researchers at the Instituto Agroforestal del Mediterraneo – Centro de Ecologia Quimica Agricola (CEQA) of the Universidad Politecnica de Valencia (Spain). The most challenging problem for the modeling of the release kinetics of this dispensers is the dificulty in obtaining experimental measurements for building the model. The procedure for obtaining these data is very costly, both time and money wise, therefore the available data across the whole season are scarce. In prior work we demonstrated the utility of using Genetic Programming (GP) for this particular problem. However, the models evolved by the GP algorithm tend to have discontinuities in those time ranges where there are not available measurements. In this work we propose the use of a multiobjective Genetic Programming for modeling the performance of the CEQA dispenser. We take two approaches, involving two and nine objectives respectively. In the first one, one of the objectives of the GP algorithm deals with how well the model fits the experimental data, while the second objective measures how “smooth” the model behaviour is. In the second approach we have as many objectives as data points and the aim is to predict each point separately using the remaining ones. The results obtained endorse the utility of this approach for those modeling problems characterized by the lack of experimental data.
Our work on modeling pheromone dispensers for sexual confusion in agricluture was presented by Anna Esparcia in the 2nd European workshop on Bio-inspired algorithms for continuous parameter optimisation, EvoNUM 2009, which is part of EvoStar 2009, the premier European event on Evolutionary Computation.
Although the presentation was scheduled at the ungodly hour of 10:30, the comments were higly favourable. As in the previous occasion where we have presented this work, the application was received with great interest, partly because computer scientists are in general unaware of where their food comes from.
However, unlike in MAEB 2009, this time we did not get what has become a memorable comment: “Us people from Madrid only go to the country to check that cows are in fact not purple”.
In such crisis times as these, the recently released chapter “Finding relevant variables in a financial distress prediction problem using Genetic Programming and Self-Organizing Maps” from volume 2 of “Natural Computing in Computational Finance” can come in quite handy.
In it we use Genetic Programming to generate models of prediction of book loses in companies, and we tackle the problem of finding the relevant variables for the prediction using Genetic Programming and Kohonen’s Self Organising Maps. This approach singificantly reduces the number of variables used for the analysis, while minimising the prediction error.
In other words, we tell you what you have to look at if you want to avoid losing money in your company – but don’t hold us responsible if you don’t like the results!
The chapter can also be obtained from SpringerLink.
You should also check our financial prediction tool PRESAGI
Our paper “Prune and Plant: A New Bloat Control Method for Genetic Programming” was presented at the Eighth International Conference on Hybrid Intelligent Systems (HIS’08) held in Barcelona the 11th-13th of September.
Here is the abstract:
This paper reports a comparison of several bloat control methods and also evaluates a new proposal for limiting the size of the individuals: a genetic operator called prune and plant. The aim of this work is to prove the adequacy of this new method. Since a preliminary study of the method has already shown promising results, we have performed a thorough study in a set of benchmark problems aiming at demonstrating the utility of the new approach. Prune and plant has obtained results that maintain the quality of the final solutions in terms of fitness while achieving a substantial reduction of the mean tree size in all four problem domains considered. In addition, in one of these problem domains prune and plant has demonstrated to be better in terms of fitness, size reduction and time consumption than any of the other bloat control techniques under comparison.