Artificial Neural Network (ANN)

clangedbivalveAI and Robotics

Oct 19, 2013 (4 years and 25 days ago)

114 views

T.W.Koh/SAK5200/20
-
12
-
2004

1

Artificial Neural Network (ANN)


Introduction to Neural Networks


ANN is an information processing paradigm that is
inspired by the way biological nervous systems,
such as the brain, process information.


It is the novel structure of the information
processing system.


It’s composed of a large number of highly
interconnected processing elements (neurons)
working in unison to solve specific problems.


T.W.Koh/SAK5200/20
-
12
-
2004

2

Artificial Neural Network (ANN)


ANNs, like people, learn by example. An An
is configured for a specific application, such
as pattern recognition through a learning
process.


Learning in biological systems involves
adjustments to the synaptic connections
that exist between the neurons.


T.W.Koh/SAK5200/20
-
12
-
2004

3

Artificial Neural Network (ANN)


Historical Background


Neural Network simulation appear to be a
recent development. However, this field
was established before the advent of
computers, and has survived at least one
major setback and several eras.


The first artificial neuron was produced in
1943 by Warren McCulloch and Walter Pits.


T.W.Koh/SAK5200/20
-
12
-
2004

4

Artificial Neural Network (ANN)


Why use Neural Networks?


Their remarkable ability to derive meaning
from complicated or imprecise data can be
used to extract patterns and detect trends
that are too complex to be noticed by
either humans or other computer
techniques.



T.W.Koh/SAK5200/20
-
12
-
2004

5

Artificial Neural Network (ANN)


A trained neural network can be thought of
as an”expert” in the category of
information it has been given to analyze.


This expert can then be used to provide
projections given new situations of interest
and answer “ what if” questions.

T.W.Koh/SAK5200/20
-
12
-
2004

6

Artificial Neural Network (ANN)


Advantages of using Neural Networks


Adaptive learning


An ability to learn how to do tasks based on the
data given for training or initial experience.


Self
-
Organization


An ANN can create its own organization or
representation of the information it receives
during learning time.


T.W.Koh/SAK5200/20
-
12
-
2004

7

Artificial Neural Network (ANN)


Real Time Operation


ANN computations may be carried out in
parallel, and special hardware devices are
being designed and manufactured which take
advantage of this capability


Fault Tolerance via Redundant Information
Coding


Partial destruction of network leads to the
corresponding degradation of performance.
However, some network capabilities may be
retained even with major network damage.

T.W.Koh/SAK5200/20
-
12
-
2004

8

Artificial Neural Network (ANN)


How the Human Brain Learns?


A typical neuron collects signals from
others through a host of fine structure
called dendrites.


The neuron sends out spikes of electrical
activity through a long, thin stand known
as an axon, which splits into thousands of
branches.

T.W.Koh/SAK5200/20
-
12
-
2004

9

Artificial Neural Network (ANN)


At the end of each branch, a structure
called a synapse converts the activity from
axon into electrical effects that inhibit or
excite activity in the connected neurons.


When a neuron receives excitatory input
that is sufficiently large compared with its
inhibitory input, it sends a spike of
electrical activity down its axon.

T.W.Koh/SAK5200/20
-
12
-
2004

10

Artificial Neural Network (ANN)


Learning occurs by changing the
effectiveness of the synapses so that the
influence of one neuron on another
changes.

Components of neuron

The synapse

T.W.Koh/SAK5200/20
-
12
-
2004

11

Artificial Neural Network (ANN)


Human Neurons to Artificial Neurons


The authors
(Christos Stergiou and Dimitrios Siganos)
conduct these neural networks by first
trying to deduce the esential features of
neurons and their interconections.


The Neuron Model

T.W.Koh/SAK5200/20
-
12
-
2004

12

Artificial Neural Network (ANN)


An Engineering Approach


A simple neuron


An artificial neuron is a device with many
inputs and one output.


The neuron has two modes of operation:


Training mode


Using mode


In the training mode, the neuron can be
trained to fire (or not), for particular input
patterns.

T.W.Koh/SAK5200/20
-
12
-
2004

13

Artificial Neural Network (ANN)


In the using mode, when a taught input pattern
is detected at the input, its associated output
becomes the current output.


If the input pattern does not belong in the
taught list of input patterns, the firing rule is
used to determine whether to fire or not.


A simple neuron

T.W.Koh/SAK5200/20
-
12
-
2004

14

Artificial Neural Network (ANN)


Firing Rules


The firing rule is an important concept in neural
networks and accounts for their high flexibility.


A firing rule determines how one calculate
whether a neuron should fire for any input
pattern.


It relates to all the input patterns, not only the
ones on which the node was trained.


A simple firing rule can be implemented by
using Hamming distance technique.

T.W.Koh/SAK5200/20
-
12
-
2004

15

Artificial Neural Network (ANN)


Examples of rules:


Attached handout….

T.W.Koh/SAK5200/20
-
12
-
2004

16

Artificial Neural Network (ANN)


References



Report:
www.doc.ic.ac.uk/Journal vol4/



Source: Narauker Dulay, Imperial College, London



Authors: Christos Stergiou and Dimitrios Siganos

T.W.Koh/SAK5200/20
-
12
-
2004

17

Artificial Neural Network (ANN)


“Neural Networks do not perform miracles. But
if used sensibly they can produce some
amazing result”


The End…


Prepared by,

T.W.Koh

T.W.Koh/SAK5200/20
-
12
-
2004

18

Coming Next..


Architecture of Neural Networks

and its learning process…