Fisica

INDICE

Abstrac................................................................................................................ 2
Introducción ....................................................................................................... 3
Objetivo general .................................................................................................4
Objetivosespecíficos...........................................................................................4
Historia de las Redes Neuronales........................................................................5
Ventajas que ofrecen las Redes Neuronales.......................................................6
Entre las ventajas se incluyen:........................................................................6Aprendizaje Adaptativo.......................................................................................7
Redes Neuronales y Computadoras Digital.........................................................7
Elementos de una Red Neuronal.........................................................................8
Unidades de Proceso: La NeuronaArtificial.........................................................9
Conexiones entre Neuronas..............................................................................11
Características de las Redes Neuronales .........................................................11
Topología de las Redes Neuronales...................................................................11
Redes de una solacapa.................................................................................11
Redes con Múltiples Capas............................................................................12
Bibliografía........................................................................................................ 13

1

Abstrac
Neural networks are just another way to emulate certain human
characteristics, such as the ability to memorize andassociate facts. If we
carefully consider those issues not can be expressed through an algorithm,
observe that all have one thing in common: experience. Man is capable
of resolve these situations coming to the experience. Thus, seems clear that
one way of approaching the problem consists in building systems that are able
to
reproduce
this
human characteristic. In short, neural networks arenothing more than an
artificial
model
simplified human brain is the most perfect example of that available for a
system that is capable of acquiring knowledge through experience. A neural
network is "a new system for information processing, whose basic unit is
processing inspired by the fundamental cell of the human nervous system:
the neuron". All processes of the human body are related in someway or
another
with the (in) activity of these neurons. They are an relatively simple man, but
when thousands of them connected together are very powerful. It is also well
known that humans are capable of learning. Learning means that those
problems can not initially resolved, can be resolved after more
information about the problem.
Therefore, Neural Networks:
•

Consist of processingunits that exchange data or information.

•
They are used to recognize patterns, including images, manuscripts and
time sequences,
financial trends.
•

Have ability to learn and improve their performance. A first classification
of neural network models could be according to their similarity with
biological reality:
1. The model of a biological nature. This includes the networks
dealingto simulate biological
neural systems and
functions hearing or some basic functions of vision.
2. The application-driven model. This model does not
have keep similarity with biological systems. Its architecture
is strongly linked to the needs of the applications for which
2

it

is designed.

Introducción
El hombre se ha caracterizado siempre por su búsqueda para mejorar sus...