Estimation of Distribution Algorithms based on Phylogenetic Trees

• Grant number: 10/01705-0
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: October 01, 2010
• End date: April 30, 2014
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Alexandre Cláudio Botazzo Delbem
• Grantee: Antonio Helson Mineiro Soares
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Associated scholarship(s): 12/17088-5 - Multiobjective estimation distribution algorithm for continuous, mixed and overlapping problems, BE.EP.DR

Abstract

The objective of the project is the development of a new Estimation of Distribution Algorithm, that is, an Evolutionary Algorithm that uses a probabilistic model of the distribution of the values of the variables of a problem to guide the search process. The main differential of the proposal is the investigation of probabilistic models used in the area of Phylogeny (that studies the reconstruction of trees - Phylogenies - which are used for represent the evolution of the species to the long of the time). Several methods exist for construction of such models in the literature. They were produced by more than a century of intense researches in this area, involving Biology and Statistics. A first Estimation of Distribution Algorithm using a Phylogenetic model was created recently by the group of research SEER of the ICMC-USP. This method, called, Phylogenetic Algorithm, uses one of the simplest phylogenetic methods from the literature. The Phylogenetic Algorithm has shown performance significantly superior in relation to the best Estimation of Distribution Algorithms. In this way, the intensification of the research on Methods of Reconstruction of Trees Phylogenetic for the developmento of Estimation of Distribution Algorithms by means of the present doctorate project should generate Phylogenetic Algorithms that are more efficient, robust and capable to solve more difficult global optimization problems, involving large-scale instances, combinatorial problems, continuous, mono- and multi-objectives problems. Thus, the proposed research should generate significant contributions in relation to the state of the art in the area that will be published in relevant journals of the area.

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

SOARES, Antonio Helson Mineiro. Estimation of distribution algorithms based on phylogenetic trees. 2014. Doctoral Thesis - Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC/SB) São Carlos.