Tesis profesional presentada por Federico Emmanuel Gómez Suárez

Licenciatura en Ingeniería en Sistemas Computacionales. Departamento de Ingeniería en Sistemas Computacionales. Escuela de Ingeniería, Universidad de las Américas Puebla.

Jurado Calificador

Presidente: Dr. Mauricio Javier Osorio Galindo
Vocal y Director: Dr. Santos Gerardo Lázzeri Menéndez
Secretario: Dra. María del Pilar Gómez Gil

Cholula, Puebla, México a 15 de diciembre de 2005.

Resumen

A Multi-Agent system, a loosely coupled network of solvers which interact to find a solution to a problem beyond individual capabilities and knowledge [29], is a common notion in the literature with application to a broad range of problems. One approach in the design and application of such systems is machine learning. Machine Learning deals with the design of programs which take advantage of data, examples and experience to improve accuracy or performance [16]. Specifically, the area of Machine Learning algorithms that deals with Multi-Agent systems is known as Ensemble Methods. Common problems addressed by these techniques are Classification and Prediction. When designing a Multi-Agent system using Ensemble Methods, 4 different stages can be identified: 1. Pre-processing 2. Partition 3. Training 4. Post-processing These 4 stages are independent and have different goals. Preprocessing refers to preparing the data in order to improve learning efficacy. Partition refers to dividing the data among the different agents. Training corresponds to the process where the data is used to learn how to solve the problem at hand. Finally post-process includes techniques to analyze or modify the learning results. At the end of this process, enough information has been learned to attempt to solve unseen problems of the same type. For each stage, there exist multiple techniques that may be applied which might be proper for one sort of problem but not for the other. One specific combination might be more adequate than other and finding an optimal combination is an aspect of our research.

To habilitate experimentation, we have designed a framework in which different interchangeable components may be connected and different Multi-Agent Systems created. These systems may then be exported through XML to be used in other applications. Using this framework, a comparative analysis on the different stages was performed, and an ensemble based solution was applied for the HLA multi-classification problem [20], for which our research represents the first attempt of applying machine learning techniques.

Table of content

Agradecimientos (archivo pdf, 12 kb)

Capítulo 1. Introduction (archivo pdf, 50 kb)

  • 1.1 Multi-Agent Systems in Decision Making problems
  • 1.2 Machine Learning and Ensemble Algorithms
  • 1.3 Objectives
  • 1.4 Research Scope

Capítulo 2. Machine Learning 8 (archivo pdf, 92 kb)

  • 2.1 Machine Learning Overview
  • 2.2 Concepts in Machine Learning
  • 2.3 Factors in Learning
  • 2.4 Machine Learning Theory

Capítulo 3. Ensemble Learning (archivo pdf, 391 kb)

  • 3.1 Problems in learning methods
  • 3.2 Ensemble Learning Design
  • 3.3 Bagging
  • 3.4 Neural Networks in bagging
  • 3.5 Boosting

Capítulo 4. Multi-Agent System Design 34 (archivo pdf, 499 kb)

  • 4.1 The Preprocess Stage
  • 4.2 The Learning Stage
  • 4.3 The Postprocess Stage

Capítulo 5. Application Area Analysis (archivo pdf, 824 kb)

  • 5.1 Experimentation focus
  • 5.2 Datasets Description

Capítulo 6. Framework Definition (archivo pdf, 1 mb)

  • 6.1 The Framework: mage
  • 6.2 The Implementation: mageImpl
  • 6.3 The Application: magegui

Capítulo 7. Experiment Result Analysis (archivo pdf, 1 mb)

  • 7.1 Benchmark comparison
  • 7.2 Preprocess Component Analysis
  • 7.3 Learning Algorithm Analysis
  • 7.4 Bagging algorithm analysis
  • 7.5 Result analysis on the HLA multi-classifier

Capítulo 8. Conclusions (archivo pdf, 23 kb)

  • 8.1 FutureWork

Referencias (archivo pdf, 32 kb)

Apéndice A. Mage Usage Example (archivo pdf, 160 kb)

Apéndice B. Digit recognition Experiment Results (archivo pdf, 28 kb)

Gómez Suárez, F. E. 2005. Automatized Multi-Agent System Design for Decision Making Problems. Tesis Licenciatura. Ingeniería en Sistemas Computacionales. Departamento de Ingeniería en Sistemas Computacionales, Escuela de Ingeniería, Universidad de las Américas Puebla. Diciembre. Derechos Reservados © 2005.