International Conference on Hybrid Artificial Intelligence Systems (HAIS) 2009

Aumentar Tamaņo del texto Disminuir Tamaņo del texto

Special Session on Evolutionary Multi-objective Machine Learning

Many current research works have combined the global search abilities of Evolutionary Computation with Machine Learning algorithms. Most of these hybrid approaches use mono-objective fitness functions. However, many issues in Machine Learning are multi-objective in nature. For instance, in feature selection, the minimization of the number of attributes and the maximization of accuracy are conflicting goals. Also, new powerful multi-objective optimization algotithms have been developed. That is why recently, multi-objective approaches have been applied to Machine Learning problems such as: improving the generalization capabilities of learning algorithms, generating diverse classifiers for building ensembles, reducing the complexity of models for improving interpretability, multi-objective-based feature selection, clustering, etc.

 ISDA 2009 Call For Papers

 

This special session welcomes articles on advances on evolutionary multi-objective-based Machine Learning. Papers comparing and studying the advantages and disadvantages of the multi-objective versus the mono-objective approach are also welcome.

Paper Submission

Papers will be reviewed by at least two members of the Program Committee. Papers accepted in the special sessions will also be published in the proceedings (Lecture Notes in Artificial Intelligence by Springer).

For submitting a paper to this special session, please visit the HAIS'09 submission web page. Please, bear in mind that there is an 8 page limit.

Contact Information

Topics

Topics include but are not limited to:

  • Evolutionary multi-objective techniques for improving the generalization capabilities of machine learning algorithms
  • Evolutionary multi-objective techniques for improving interpretability of models
  • Evolutionary multi-objective feature selection
  • Evolutionary multi-objective ensemble generation
  • Empirical and/or theoretical comparisons between evolutionary mono-objective and multi-objective machine learning techniques
  • Multi-objective Genetic Programming
  • New evolutionary multi-objective algorithms speciallized in machine learning
  • Applications of evolutionary multi-objective learning

Co-Chairs at EVANNAI

  • Ricardo Aler
  • Inés M. Galván
  • José M. Valls

PC Members

  • Henrik Bostrom, School of Humanities and Informatics, University of Skövde. Sweden
  • Juan Carlos Fernandez Caballero, Computer Science and Numerical Analysis Department , Universidad de Córdoba. Spain
  • César Hervás, Computer Science and Numerical Analysis Department , Universidad de Córdoba. Spain
  • Andrew Hunter, Department of Computing and Informatics, University of Lincoln, UK
  • Pedro Isasi, Computer Science Department, Universidad Carlos III de Madrid. Spain
  • Yaochu Jin, Honda Research Institute Europe / Bielefeld University. Germany
  • David Quintana, Computer Science Department, Universidad Carlos III de Madrid. Spain
  • Peter Rockett, Department of Electronic Engineering, University of Sheffield. UK
  • Katya Roríguez-Vázquez, Ingeniería de Sistemas Computacionales y Automatización, UNAM,México.
  • El-Ghazali Talbi, INRIA Futurs / University of Lille. France
  • Yago Sáez, Computer Science Department, Universidad Carlos III de Madrid. Spain