Curso de Pós-Graduação em Ciência da Computação

INE 6103 - Redes Neurais

INDEX

Objetivos

Súmula

Bibliografia

Trabalhos

Slides

Leituras Selecionadas

Lista de Exercícios

Links Interessantes

Software

Objetivos

Oferecer ao aluno uma introdução à abordagem da IA conexionista, descrevendo características de funcionamento, formas de aprendizado e aplicações típicas. Vários modelos de redes serão estudados, seguindo-se o uso de softwares de simulação.

TOP

Súmula

Introdução - Inteligência Artificial Simbólica e Conexionista.
Histórico das Redes Neurais Artificiais.
Nomenclatura Básica, o Neurônio Biológico e Revisão de Álgebra Linear.
Tipos de Redes, Arquiteturas, Características e Aplicações.
Modelos Básicos: Perceptron, Adaline, Madaline.
Modelos Básicos: Multi-layer Perceptron e Regra de Aprendizado Back-Propagation.
Redes “Counterpropagation”.
Memórias Associativas: Hopfield e BAM.
Redes de Kohonen.
Redes ART
Redes Recorrentes.

TOP

Bibliografia

Freeman, James A. & Skapura, David M. Neural Networks: Algorithms, Applications and Programming Techniques. Addison-Wesley Publishing, 1992.
Haykin, Simon. Neural Networks: a comprehensive foundation. IEEE Press, 1994.
Barreto, Jorge M.Inteligência Artificial: No limiar do século XXI. ??? Edições, 1997
Kovács, Zsolt Laszlo. Redes Neurais Artificiais: fundamentos e aplicações. Collegium Cognitio, 1997.
Rumelhart, D.; Hinton, G. & Williams, R. Learning Internal representation by Error Propagation. In: Parallel Distributed Processing: explorations in the microstructure of cognition - Vol 1. MIT Press, 1986.
Arbib, Michael A. (Ed). The handbook of Brain Theory and Neural Networks. MIT Press, 1995.

TOP

Slides

Curso de Redes Neurais
Introdução, Objetivo, Súmula e Bibliografia
Origens da IA
Duas abordagens de IA: Simbólica e Conexionista
Fases Históricas
Computação baseada em: Instruções e Neural
IA Simbólica x IA Conexionista
Interesses em usar RNA
Novo paradigma de programação
Fundamentações biológicas
Uma breve revisão de Álgebra Linear
RNA - Origens
Circuitos neurais e computação
Modelo de neurônio artificial
O Período negro
O Renascimento
RNA - Tempos Modernos
Elementos Básicos de um Neurônio Artificial
Topol. de RNA - Como os Neurônios se Conectam
Aprendizado de RNA
Algoritmo Backpropagation
RNA - Memórias Associativas
RNA - Memórias BAM
RNA - Memórias Associativas - Memória Hopfield
RNA - A Rede Counterpropagation
RNA - Redes Auto-Organizáveis
RNA - Redes ART
Full Slide em PB Download

TOP

Leituras Selecionadas

Barreto,Jorge Muniz, Inteligência Artificial no limiar do século XXI,Florianópolis, Duplic, 2a. edição, 2000.

CEREBRO E A SUA MENTE, O
UMA INTRODUÇAO A NEUROCIENCIA COMPUTACIONAL
Autor: KOVACS, ZSOLT L
Editora: EDIÇAO DO AUTOR
Assunto: CIENCIAS COGNITIVAS
Ficha Técnica:
Livro em PORTUGUES - ENCAD. C/ SOBRECAPA - 1 Edição 1997 - 241 pág.

REDES NEURAIS ARTIFICIAIS
FUNDAMENTOS E APLICAÇOES
Autor: KOVACS, ZSOLT L
Editora: COLLEGIUM COGNITIO
Assunto: INFORMATICA-INTELIGENCIA ARTIFICIAL
Ficha Técnica:
ISBN 8586396028 - Livro em PORTUGUES - ENCADERNADO - 2 Edição 1996 - 174 pág.

PERCEPTRONS - EXPANDED EDITION
AN INTRODUCTION TO COMPUTATIONAL GEOMETRY
Autor: PAPERT, SEYMOUR
Autor: MINSKY, MARVIN L.
Editora: MIT PRESS
Assunto: INFORMATICA

Perceptrons--the first systematic study of parallelism in computation--has remained a classical work on threshold automata networks for nearly two decades. It marked a historical turn in artificial intelligence and is required reading for anyone who wants to understand the connectionist revival that is going on today. Artificial intelligence research, which for a time concentrated on the programming of von Neumann computers, is swinging back to the idea that intelligence might emerge from the activity of networks of neuron-like entities. Minsky and Papert's book was the first example of a mathematical analysis carried far enough to show the exact limitations of a class of computing machines that could seriously be considered as models of the brain. Now the new developments in mathematical tools, the recent interest of physicists in the theory of disordered matter, the new insights into and psychological models of how the brain works, and the evolution of fast computers that can simulate networks of automata have given Perceptrons new importance. Minsky and Papert have added a chapter to their seminal study in which they discuss the current state of parallel computers, review developments since the appearance of the 1972 edition, and identify new research directions related to connectionism. The central theoretical challenge facing connectionism, they observe, is in reaching a deeper understanding of how 'objects' or 'agents' with individuality can emerge in a network. Progress in this area would link connectionism with what the authors have called 'society theories of mind'.

Ficha Técnica:
ISBN 0262631113 - Livro em INGLES - BROCHURA - 1988

Neural Networks : Algorithms, Applications, and Programming Techniques (Computation and Neural Systems Series)
by James A. Freeman, David M. Skapura

ASIN: 0201513765

Editorial Reviews
From Book News, Inc.
A textbook for a graduate or advanced undergraduate course in neural networks for computer-science or engineering students. Presumes the standard calculus, differential equations, and advanced mathematics acquired in the first three years of an engineering curriculum. Includes models inspired by, but not found by, studies of the brain. Annotation copyright Book News, Inc. Portland, Or.

Ingram
Freeman and Skapura provide a practical introduction to artificial neural systems (ANS). The authors survey the most common neural-network architectures and show how neural networks can be used to solve actual scientific and engineering problems and describe methodologies for simulating neural-network architectures on traditional digital computing systems.

REDES NEURAIS - PRINCIPIOS E PRATICAS
Autor: HAYKIN; SIMON