R. Aler - Academia.edu (original) (raw)

Papers by R. Aler

Intelligent Agent Software Engineering, 2003

In this paper, we present a multiagent system approach with the purpose of building computer prog... more In this paper, we present a multiagent system approach with the purpose of building computer programs. Each agent in the multiagent system will be in charge of evolving a part of the program, which in this case, can be the main body of the program or one of its different subroutines. There are two kinds of agents: the manager agent and the genetic programming (GP) agents. The former is in charge of starting the system and returning the results to the user. The GP agents include skills for evolving computer programs, based on the genetic programming paradigm. There are two sorts of GP agents: some can evolve the main body of the program and the others evolve its subroutines. Both kinds of agents cooperate by telling each other their best results found so far, so that the search for a good computer program is made more efficient. In this paper, this multiagent approach is presented and tested empirically.

Archivos de Zootecnia

La comparación entre la curva de producción real del huevo y la gráfica propuesta por las pautas ... more La comparación entre la curva de producción real del huevo y la gráfica propuesta por las pautas de gestión, tiene como objetivo la evaluación continua del rendimiento. Los objetivos de este estudio fueron comparar la capacidad de la curva de ajuste de la producción diaria de huevo de Lokjorst (LM), la red neuronal del perceptrón multicapa (MP) y las redes neuronales recurrantes de Jordania y Elman (RNNJ y RNNE, respectivamente) para la predicción del huevo diario producción en gallinas ponedoras comerciales. Los modelos se instalaron utilizando 4650 datos de 12 lotes seleccionados. Los modelos MP y LM dieron un buen ajuste a los datos, con valores de correlación superiores a 0,95 y que representan más del 95% de la variabilidad en la producción diaria de óvulos. Para el pronóstico de producción, MP fue una técnica con una precisión aceptable y menos variación. El modelo MP se recomienda como herramienta de ajuste y previsión de la curva diaria de producción de huevos en gallinas co...

Concurrency and Computation: Practice and Experience

This article addresses two issues in solar energy forecasting from the numerical weather predicti... more This article addresses two issues in solar energy forecasting from the numerical weather prediction (NWP) models using machine learning. First, we are interested in determining the relevant information for the forecasting task. With this purpose, a study has been carried out to evaluate the influence on accuracy of the number of NWP grid nodes used as input for the forecasting model, as well as their relative importance. Several machine learning (support vector machines and gradient boosting) and feature selection algorithms (linear, ReliefF, and local information analysis) have been used in this study. The second aim is to be able to predict solar energy for locations where no previous production data are available. To address this goal, an approach consisting on modeling regions in the grid is proposed. Models (aggregate models) use as input attributes the meteorological variables relevant for the region and two new inputs to identify the location of each station: the latitude and the longitude. Those models can be used to predict energy production for existing stations and for new locations, represented by latitude and longitude. Copyright © 2015 John Wiley & Sons, Ltd.

IEEE Latin America Transactions, 2008

Archivo abierto el Repositorio Institucional de la Universidad Carlos III, E-Archivo Instituciona... more Archivo abierto el Repositorio Institucional de la Universidad Carlos III, E-Archivo Institucional Repository of University Carlos III.

Journal of Geophysical Research: Atmospheres

AI Magazine

In this article, we discuss the problem of transferring search heuristics from one planner to ano... more In this article, we discuss the problem of transferring search heuristics from one planner to another. More specifically, we demonstrate how to transfer the domain-dependent heuristics acquired by one planner into a second planner. Our motivation is to improve the efficiency and the efficacy of the second planner by allowing it to use the transferred heuristics to capture domain regularities that it would not otherwise recognize. Our experimental results show that the transferred knowledge does improve the second planner's performance on novel tasks over a set of seven benchmark planning domains.

Resumen Este trabajo presenta un sistema de aprendizaje automati- co, HHamlet, para un planificad... more Resumen Este trabajo presenta un sistema de aprendizaje automati- co, HHamlet, para un planificador h´ibrido, Hybis, que mezcla plani- ficacion jerarquica HTN (Hierachical Task Network) y planificacion de orden parcial (POP). HHamlet esta basado en Hamlet, un sistema multi-estrategia deductivo-inductivo, que aprende reglas de control para el planificador no lineal de orden total Prodigy4.0. Los dominios en Hy- bis representan

Proceeding of the fifteenth annual conference on Genetic and evolutionary computation conference - GECCO '13, 2013

2001 IEEE International Conference on Systems, Man and Cybernetics. e-Systems and e-Man for Cybernetics in Cyberspace (Cat.No.01CH37236), 2001

Solving problems in dynamic and heterogeneous environments where information sources change its f... more Solving problems in dynamic and heterogeneous environments where information sources change its format representation and the data stored along the time is a very complex problem. In previous work we have presented a system called MAPWeb (MultiAgent Planning in the Web) that tries to solve those problems by integrating artificial intelligence planning techniques within the MultiAgent framework. Basically, MAPWeb allows cooperative work between planning agents and Web agents. The purpose of MAP Web is to find solutions to travel problems. In order to give detailed solutions, MAP Web uses information gathering techniques to retrieve travel information that is made available by many different companies. However, Web access to the information sources is quite time expensive. In this paper, we try to minimize the number of Web queries by using caching techniques based on relational databases. Experimental results show that the reduction in Web access time is quite important, while maintaining the number of solutions found.

Lecture Notes in Computer Science, 2004

Proceedings. International Conference on Machine Learning and Cybernetics, 2002

In this paper we present a learning method that is able to automatically acquire control knowledg... more In this paper we present a learning method that is able to automatically acquire control knowledge for a hybrid HTN• POP planner called HYBIS. HYBIS decomposes a problem in subproblems using either a default method or a user-defined decomposition method. Then, at each level of abstraction, it generates a plan at that level using a POCL (Partial Order Causal Link) planning technique. Our learning approach builds on HAMLET, a system that learns control knowledge for a total order non•linear planner (PRODIGY 4.0). In this paper, we focus on the operator selection problem for the POP component of HYBIS, which is very important for efficiency purposes.

Lecture Notes in Computer Science, 2009

Many classification algorithms use the concept of distance or similarity between patterns. Previo... more Many classification algorithms use the concept of distance or similarity between patterns. Previous work has shown that it is advantageous to optimize general Euclidean distances (GED). In this paper, data transformations are optimized instead. This is equivalent to searching for GEDs, but can be applied to any learning algorithm, even if it does not use distances explicitly. Two optimization techniques have been used: a simple Local Search (LS) and the Covariance Matrix Adaptation Evolution Strategy (CMA-ES). CMA-ES is an advanced evolutionary method for optimization in difficult continuous domains. Both diagonal and complete matrices have been considered. Results show that in general, complete matrices found by CMA-ES either outperform or match both Local Search, and the classifier working on the original untransformed data.

Lecture Notes in Computer Science, 2001

Evolutionary based learning systems have proven to be very powerful techniques for solving a wide... more Evolutionary based learning systems have proven to be very powerful techniques for solving a wide range of tasks, from prediction to optimization. However, in some cases the learned concepts are unreadable for humans. This prevents a deep semantic analysis of what has been really learned by those systems. We present in this paper an alternative to obtain symbolic models from subsymbolic learning. In the first stage, a subsymbolic learning system is applied to a given task. Then, a symbolic classifier is used for automatically generating the symbolic counterpart of the subsymbolic model. We have tested this approach to obtain a symbolic model of a neural network. The neural network defines a simple controller of an autonomous robot. A competitive coevolutive method has been applied in order to learn the right weights of the neural network. The results show that the obtained symbolic model is very accurate in the task of modelling the subsymbolic system, adding to this its readability characteristic.

2007 IEEE Congress on Evolutionary Computation, 2007