• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.43.2022.tde-22022023-134919
Document
Author
Full name
Pietro Zanin
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2022
Supervisor
Committee
Alfonso, Nestor Felipe Caticha (President)
Neirotti, Juan Pablo
Stariolo, Daniel Adrián
Title in Portuguese
Análise de redes neurais de atratores interagentes por meio de um modelo com solução analítica
Keywords in Portuguese
Algoritmos de Aprendizado
Mecânica Estatística
Método de réplicas
Modelos de agentes
Redes Neurais de Atratores
Abstract in Portuguese
Neste trabalho construímos e analisamos um modelo com o objetivo de ampliar ideias do algoritmo de unlearning para tentar entender redes neurais interagentes. Nos baseamos não só em vários trabalhos de redes neurais que giram em torno desta ideia, mas também em alguns artigos de ciências sociais relacionados. O modelo é construído introduzindo uma modificação do Hamiltoniano de outros modelos, no qual introduzimos uma interação entre diferentes redes. Mudar a magnitude desta interação leva a resultados diferentes, sendo eles não triviais e ricos. Em geral, discutimos em quais regiões essa interação é benéfica e de que maneira ela pode ser benéfica. Apesar do modelo ser complexo demais para ser comparado com dados reais, ele apresenta comportamentos qualitativos que mimetizam algumas dinâmicas sociais de maneira interessante. Além disso, o modelo também é de interesse para a área de redes neurais, pois mostra uma maneira em que redes podem ser melhoradas significativamente de maneira eficiente.
Title in English
Analysis of interacting attractor neural networks with a model with analytic solution
Keywords in English
agent models
attractor neural networks
learning algorithms
replica method
Statistical Mechanics
Abstract in English
In this work we built and analyzed a model with the goal of enlarging and generalizing ideas from the unlearning algorithm to try to understand interacting neural networks. Besides basing this work in previous neural networks articles related to that subject, we also used some social science articles to further develop the work. The model is built by introducing a change in the Hamiltonian of other models, in which we introduce an interaction between different networks. Changing the magnitude of this interaction leads to different results, which are interesting and non-trivial. In general, we try to understand the regions where this interaction is beneficial and in which ways it can be beneficial. Despite the model being too complex to be compared to real data, it shows qualitative behaviors that mimic some social dynamics in an curious way. Additionally, the model is also interesting to the area of neural networks, as it shows a way to train the networks in an efficient way.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
dissPietroZanin.pdf (2.23 Mbytes)
Publishing Date
2023-03-20
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.