Kazan National Research Technical University named after A.N. Tupolev German-Russian Institute of Advanced Technologies (GRIAT)
Table of contents
References
Questions for motivation discussion
Types of learning
Artificial Neural Network
Artificial Neural Networks
ANN vs Computers
ANN vs Computers
Biological neuron
Biological neuron
Artificial Neuron Structure
Common activation functions
Examples of ANN topologies
Fundamentals of learning and training samples
Fundamentals of learning and training samples
Fundamentals of learning and training samples
Fundamentals of learning and training samples
Fundamentals of learning and training samples
Fundamentals of learning
General learning procedure
Using training samples
Learning curve
Error measurement
When do we stop learning?
Using neural networks in practice (discussion)
Single layer network with binary threshold activation function
Single layer network with binary threshold activation function
Practice with single layer neural network
Hebbian learning rule
Hebbian learning rule (matrix form)
Practice with hebbian learning rule
Delta rule (Widrow-Hoff rule)
Delta rule (Widrow-Hoff rule)
Delta rule (Widrow-Hoff rule)
Delta rule algorithm
Linear classifiers
Practice with delta rule
Rosenblatt's single layer perceptron
Rosenblatt's single layer perceptron
Rosenblatt's learning algorithm
Rosenblatt's single layer perceptron
Practice with Rosenblatt's perceptron
Associative memory
Associative memory
Autoassociative memory based on sign activation function
Practice with autoassociative memory
Using single layer neural networks for time series forecasting
Using single layer neural networks for time series forecasting
Practice with time series forecasting
Multilayer perceptron
Multilayer perceptron
Multilayer perceptron
Multilayer perceptron
Classification ability
Classification ability
Classification ability
Backpropagation algorithm
Basic steps
Backpropagation
Backpropagation
Backpropagation
Backpropagation
Backpropagation
Backpropagation algorithm
Backpropagation algorithm
Practice. Calculation delta-rule expressions for various activation functions
Some problems
Some problems
Practice with multilayer perceptron
Recurrent neural networks
Hopfield network
Hopfield network
Hopfield network as associative memory
Using hopfield network as associative memory
Hopfield network as associative memory
Example
Practice with Hopfield network
Hamming network
Hamming network
Hamming network working algorithm
Self-organizing maps
Self-organizing maps
Self-organizing maps (SOM)
Self-organizing maps
Scheme of training of self-organizing map
Competitive learning
Competitive learning
Vector quantization
Vector quantization
Kohonen Maps
Kohonen maps
Kohonen maps learning procedure
Training and Testing
Training
Training and Verification
Verification
Summary (Discussion)
Summary
Questions and Comments
Thank you for attention

Neural networks

1. Kazan National Research Technical University named after A.N. Tupolev German-Russian Institute of Advanced Technologies (GRIAT)

NEURAL NETWORKS
by Dr. Igor Anikin

2. Table of contents

The basic concepts of neural networks
Single layer neural networks
Artificial neural networks.
The structure of an artificial neuron.
Activation functions.
Basic paradigms of neural networks.
Fundamentals of learning and training samples.
Using neural networks in practice
Rosenblatt's single layer perceptron.
Learning single layer neural networks.
Associative memory and its realization on single layer neural networks.
Using single layer neural networks for pattern recognition and time
series forecasing.
Multilayer perceptrons
The structure of multilayer perceptrons
Back propagation of error.
Using multilayer perceptrons for pattern recognition and time series
forecasing.

3.

Self-organizing maps
The principle of unsupervised learning.
Kohonen self-organizing maps.
Learning Kohonen networks.
Practical using of Kohonen networks
Recurent neural networks
Neural networks with feedback.
Hopfield neural network.
Hamming neural network.
Training Hopfield and Hamming neural networks.
Practical using of Hopfield and Hamming neural networks.
Training and Testing
Training error and testing error.

4. References

1.
2.
3.
4.
David Kriesel.
Neural
A
brief Introduction
networks
to
//
http://www.dkriesel.com/en/science/neural_net
works
Raul Rojas. Neural Networks. A Systematic
Introduction
//
http://www.inf.fu-berlin.de/inst/ag-ki/rojas_
home/documents/1996/NeuralNetworks/neuron.pdf
.
L.P.J. Veelenturf. Analysis and Application of
Artificial
Neural
Networks
//
http://www.ru.lv/~peter/zinatne/ebooks/Anal
ysis%20and%20Applications%20of%20Artificial
%20Neural%20Networks.pdf
Artificial Neural Networks – Methodological
Advances and Biomedical Applications //

5.

The basic concepts of
neural networks

6. Questions for motivation discussion

What tasks are machines good at doing that
humans are not?
What tasks are humans good at doing that
machines are not?
What tasks are both good at?
What does it mean to learn?
How is learning related to intelligence?
What does it mean to be intelligent?
Do you believe a machine will ever been
intelligent?
If a computer were intelligent, how would
you know?

7. Types of learning

Knowledge acquisition from expert.
Knowledge acquisition from data:
Supervised learning – the system is supplied
with a set of training examples consisting of
inputs and corresponding outputs, and is
required to discover the relation or mapping
between them.
Unsupervised learning – the system is
supplied with a set of training examples
consisting only of inputs. It is required to
discover what appropriate outputs should be.

8. Artificial Neural Network

An extremely simplified model of the
human’s brain
Transforms inputs into the best outputs
(some neural networks are the universal
function approximators).

9. Artificial Neural Networks

Development of Neural Networks date back to the early
1940s.
It experienced an upsurge in popularity in the late 1980s
due to discovery of new techniques of NN training.
Some NNs are models of biological neural networks and
some are not, but historically, much of the inspiration for the
field of NNs came from the desire to produce artificial
systems capable of sophisticated, perhaps intelligent,
computations similar to those that the human brain
routinely performs, and thereby possibly to enhance our
understanding of the human brain.
Most NNs have some sort of training rule. In other words,
NNs learn from the examples (as children learn to recognize
dogs from examples of dogs) and exhibit some capability for
generalization beyond the training data.

10. ANN vs Computers

Computers
have
programmed
to
be
explicitly
Analyze the problem to be solved.
Write the code in a programming language.
Neural networks learn from the examples
No requirement of an explicit description of the problem.
No need for a programmer.
The neural computer adapts itself during a training
period, based on examples of similar problems even
without a desired solution to each problem. After
sufficient training the neural computer is able to relate
the problem data to the solutions, inputs to outputs, and
it is then able to offer a viable solution to a brand new
problem.

11. ANN vs Computers

Digital Computers
Deductive Reasoning. We
apply known rules to input
data to produce output.
Computation
is
centralized, synchronous,
and serial.
Memory is literally stored,
and location addressable.
Not fault tolerant. One
transistor goes and it no
longer works.
Exact.
Static connectivity.
Applicable if well-defined
rules
accessible
with
precise input data.
Neural Networks
Inductive
Reasoning. We use
given input and output data
(training examples) to make a
reasoning.
Computation
is
collective,
asynchronous, and parallel.
Memory
is
distributed,
internalized, short term and
content addressable.
Fault tolerant, redundancy, and
sharing of responsibilities.
Inexact.
Dynamic connectivity.
Applicable if rules are unknown
or complicated, or if data are
noisy or partial.

12. Biological neuron

13. Biological neuron

Many
“neurons”
co-operate
perform the desired function
Basic elements:
Axon
Dendrite
Synapse
to

14. Artificial Neuron Structure

The output of a neuron is a function of the
weighted sum of the inputs plus a bias
n
S x j w j b,
j 1
y f (S )

15. Common activation functions

16.

17. Examples of ANN topologies

Single layer ANN
Multilayer ANN
ANN with one recurrent layer

18. Fundamentals of learning and training samples

The weights in a neural network are the
most important factor in determining its
function.
A training set is a set of training patterns,
which we use to train our neural net.
Training is the act of presenting the
network with some sample data and
modifying
the
weights
to
better
approximate the desired function

19. Fundamentals of learning and training samples

There are two main types of training
Supervised Training
Supplies the neural network with inputs and the
correct outputs (results).
We can estimate a error vector for certain input.
Response of the network to the inputs is measured.
The weights are modified to reduce the difference
between the actual and desired outputs
Unsupervised Training
The training set only consists of input patterns.
The neural network adjusts its own weights so that
similar inputs cause similar outputs. The network
identifies the patterns and differences in the inputs
without any external assistance

20. Fundamentals of learning and training samples

A training pattern is an input vector p
with the components x1, x2, . . . , xn
whose desired output is known.
By entering the training pattern into
the network we receive an output that
can be compared with the desired
output.
The set of training patterns is called P.
It contains a finite number of ordered
pairs (p, t) of training patterns with
corresponding desired output t.

21. Fundamentals of learning and training samples

Teaching input. Let j be an output neuron.
The teaching input tj is the desired and
correct value j should output after the input of
a certain training pattern.
Analogously to the vector p the teaching
inputs t1, t2, . . . , tn of the neurons can also be
combined into a vector t. This vector always
refers to a specific training pattern p and
contained in the set P of the training patterns.

22. Fundamentals of learning and training samples

Error vector. For several output neurons
Ω1,Ω2, . . . ,ΩO the difference between
output vector and teaching input under a
training input p is referred to as error
vector.
t1 y1
E p ...
t y
O
O

23. Fundamentals of learning

Let P be the set of training patters. In
learning procedure we realize finite
number of iterations or epochs.
Epoch – single presentation of the
entire data to the neural network.
Typically many epochs are required to
train the neural network
Iteration - the process of providing the
network with an single input and
updating the network's weights

24. General learning procedure

Let P be the set of n training patters pn
For i=1 to n
begin
1.
2.
We calculate NN output vector yi for the training
pattern pi.
We compare yi with desired output ti. Then we
calculate the error of output and make modification
of weights.
end
3.
If total error for the training set P more
than some threshold then go to the step 2

25. Using training samples

We have to divide the set of training samples
into two subsets:
one training set really used to train;
one verification set to test our progress of learning.
The usual division relations are, 70% for
training data and 30% for verification data
(randomly chosen).
We can finish the training process when the
network provides the good results on the
training data as well as on the verification
data.

26. Learning curve

The learning curve indicates the progress
of the error, which can be determined in
various ways. This curve can indicate
whether the network is progressing or not.

27. Error measurement

Let Ω be the output neuron and O be the
set of output neurons.
The specific error Errp is based on a single
training sample.
The total error Err is based on all training samples.
Err
Errp
p P

28. When do we stop learning?

Generally, the training process is
stopped when the user in front of the
learning computer "thinks" the error is
small enough.

29. Using neural networks in practice (discussion)

Classification
in marketing: consumer spending pattern classification
In defence: radar and sonar image classification
In medicine: ultrasound and electrocardiogram image
classification, EEGs, medical diagnosis
Recognition and identification
In general computing and telecommunications : speech, vision and
handwriting recognition
In finance: signature verification and bank note verification
Assessment
In engineering: product inspection monitoring and control
In defence: target tracking
In security: motion detection, surveillance image analysis and
fingerprint matching
Forecasting and prediction
In finance: foreign exchange rate and stock market forecasting
In agriculture: crop yield forecasting
In marketing: sales forecasting
In meteorology: weather prediction

30.

Single layer neural networks

31. Single layer network with binary threshold activation function

n
y j F S j F wij xi T j
i 1
w11
w21
W
...
w
n1
Matrix form
S W T X T
w12
w22
...
wn 2
... w1m
... w2 m
... ...
... wnm

32. Single layer network with binary threshold activation function

1, S 0
y
0, S 0
S w11 x1 w21 x2 T1
w11 x1 w21 x2 T1 0
x2
T1 w11
x1
w21 w21

33. Practice with single layer neural network

1.
2.
Performing a calculations in single layer
neural networks with using direct and matrix
form. Using various activation functions.
Using single layer neural networks with
binary threshold activation function as linear
classifier. Adjusting the linear classifier
based on training samples.

34. Hebbian learning rule

-Introduced by Donald Hebb in his 1949 book “The Organization
of Behavior”.
-Describes a basic mechanism for synaptic plasticity
wij t 0 0, i, j
wij t 1 wij t xi y j , where t time
S w11 x1 w21 x2 T
w11 t 1 w11 t x1 y1
w21 t 1 w21 t x2 y1
T t 1 T t y1

35. Hebbian learning rule (matrix form)

x11
2
x1
X
...
L
x1
x12
x22
...
x2L
... x1n
... xn2
, where X i x1i ,..., xni input pattern
... ...
L
... xn
S XW
Y F (S )
W X T Y hebbian learning rule

36. Practice with hebbian learning rule

Construction the neural network based
on hebbian learning rule for modeling
OR logical operator

37. Delta rule (Widrow-Hoff rule)

1. The delta rule is a gradient descent learning rule for
updating the weights of the inputs to artificial
neurons in single-layer neural network
2. The goal is to minimize the error between the actual
outputs and the target outputs in the training data
3. For each (input/output) training pair, the delta rule
determines the direction you need to adjust wij to
reduce the error for that training pair.
4. Derivatives are used for teaching

38. Delta rule (Widrow-Hoff rule)

ADALINE (ADAptive LINear Element) network
n
y1 w j1 x j T
j 1
L
1 L k k 2
E E (k ) ( y1 t )
2 k 1
k 1
1
E (k ) ( y1k t k ) 2
2

39. Delta rule (Widrow-Hoff rule)

Gradient descent method: find
the steepest
way
down
the
slope from where you are, and
take a step in that direction
E (k )
w j1 (t 1) w j1 (t )
w j1 (t )
E
E y1k
k
( y1k t k ) x kj
w j1 (t ) y1 w j1
E (k )
T (t 1) T (t )
T (t )
E
E y1k
k
( y1k t k )
T (t ) y1 T

40. Delta rule algorithm

1.
2.
3.
4.
5.
Define 0<a<1 and Emin
Initialize the weights with some
small random value
Take input pattern and calculate
output vector.
Modify weights and bias according
delta rule.
Do steps 3-4 until E<Emin

41. Linear classifiers

42. Practice with delta rule

Construction the ADALINE neural
network
(linear
classifier
with
minimum error value) based on given
training patterns.

43. Rosenblatt's single layer perceptron

The perceptron is an algorithm for
supervised classification of an input
into one of several possible nonbinary outputs.
It is a type of linear classifier.
Was invented in 1957 by Frank
Rosenblatt as a machine for image
recognition.

44. Rosenblatt's single layer perceptron

1, s 0
f (s)
1, s 0
Learning rule

45. Rosenblatt's learning algorithm

1.
2.
3.
4.
5.
Initialise the weights and the threshold.
Weights may be initialised to 0 or to a
small random value.
Take input pattern x from X and
calculate output vector y from Y.
If yi=tj then wij will not change.
If yi≠tj then wij(t+1) = wij (t) + xi tj
Do steps 2-4 until yi=tj for whole
training set

46. Rosenblatt's single layer perceptron

It
was
quickly
proved
that
perceptrons could not be trained to
recognize many classes of patterns.
It is linear classifier. For example, it
is impossible for these classes of
network to learn an XOR function.

47. Practice with Rosenblatt's perceptron

Construction the linear classifier (Rosenblatt’s neural
network perceptron) based on given training patterns.

48. Associative memory

Associative memory (computer science) - a
data-storage device in which a location is
identified by its informational content rather
than by names, addresses, or relative
positions, and from which the data may be
retrieved. This memory enable one to retrieve
a piece of data from only a tiny sample of itself.
Associative memory (psychology) - recalling
a previously experienced item by thinking of
something that is linked with it, thus invoking
the association

49. Associative memory

Autoassociative
memories
are
capable of retrieving a piece of data
upon presentation of only partial
information from that piece of data
Heteroassociative
memories
can
recall an associated piece of datum
from one category upon presentation
of data from another category.

50. Autoassociative memory based on sign activation function

Neural
structure:
Activation function
network
1, s 0
f (s)
1, s 0
Number of neurons
in the input layer =
Number of neurons
in the output layer
Learning rule
(adopted hebbian rule)
W XT X
Example:

51. Practice with autoassociative memory

Realization of the associative memory
based on sign activation function.
Working with multiple patterns.
Recognition of the original and noisy
patterns.
Investigation of the properties and
constraints of the associative memory
based on sign activation function.

52. Using single layer neural networks for time series forecasting

A time series points, measured
in time spaced
intervals
sequence of data
typically at points
at uniform time

53. Using single layer neural networks for time series forecasting

Training samples
x ( 2)
x(1)
x(3)
x ( 2)
X
...
...
x(m p ) x(m p 1)
x( p )
... x( p 1)
...
...
... x(m 1)
...
x( p 1)
x( p 2)
Y
...
x ( m)

54. Practice with time series forecasting

Using ADALINE neural networks for
currency forecasting:
Creation the training set from the raw
data (www.val.ru).
Learning the ADALINE.
Training ADALINE network with using
delta rule and estimation the error.

55. Multilayer perceptron

56. Multilayer perceptron

A multilayer
perceptron (MLP)
is
a  feed
forward artificial neural network model that maps
sets of input data onto a set of appropriate
outputs.
Consists of multiple layers (input, output, one or
several hidden layers) of nodes in a directed
graph, with each layer fully connected to the next
one.
Neurons with a nonlinear activation function.
Utilizes
a supervised
learning technique
called backpropagation of error.
Typical structure

57. Multilayer perceptron

Structure (2 hidden layers)
Calculation the output Y for input vector X

58. Multilayer perceptron

Activation function is not a threshold
Function approximator
Usually a sigmoid function
Not limited to linear problems
Information flows in one direction
The outputs of one layer act as inputs to
the next layer

59. Classification ability

A single layer network can only find
a linear discriminant function.
It can divide the input space by
means of hyperplane (straight lines
in two-dimensional space)

60. Classification ability

Universal Function Approximation Theorem
MLP with one hidden layer can approximate
arbitrarily closely every continuous function that
maps intervals of real numbers to some output
interval of real numbers
f:[0,1]n->[0,1]
2n+1 neurons in hidden layer.
 Can form single convex
decision regions
One hidden layer is sufficient
for the large majority of problems

61. Classification ability

Any function can be approximated to arbitrary
accuracy by a network with two hidden layers
MLP with two hidden layers can classify sets of
any form. It can form arbitrary disjoint decision
regions

62. Backpropagation algorithm

D. Rumelhart, G. Hinton, R. Williams (1986)
Most common method of obtaining the
weights in the multilayer perceptron
A form of supervised training
The basic backpropagation algorithm is
based on minimizing the error of the
network using the derivatives of the error
function
Backpropagation of error generalizes the
delta rule

63. Basic steps

Forward propagation of a training
pattern's input through the neural
network in order to generate the
propagation's output activations.
Backward propagation of the
output’s error through the neural
network using the training pattern
target in order to generate the deltas
of all output and hidden neurons.

64. Backpropagation

65. Backpropagation

We use gradient descent method for
minimizing the error

66. Backpropagation

Theorem. For any hidden layer i of the neural
network, error of the neuron i calculates by
recursive way through the errors of neurons of
the next layer j.
m
i j F ( S j ) wij
j 1
where m – number of neurons in the next layer j
wij – weights between neuron i and neurons in the
next layer j
Sj – weighted sum for the neuron j in next layer.
Proof

67. Backpropagation

Theorem. We can calculate derivatives of error
E through the weights w and bias T by
following way.
Proof

68. Backpropagation

Backpropagation rule

69. Backpropagation algorithm

the training speed (0< <1) and
desired minimal error Em
2.Initialize the weights and biases by random
way.
3.Take consequently all input patterns x from X.
1.Define
y j F ( vector
wij yi T j ) y by following way
Calculate output
i
Realize backpropogation shceme by following
way
ij
j
i
Modify ijweights
and j biases
by following way
w (t 1) w (t ) F ( S ) y
T j (t 1) T j (t ) j F ( S j )

70. Backpropagation algorithm

4. Calculate overall error for all patterns
1 L
E ( y kj t kj ) 2
2 k 1 j
5. If E>Em then go to the step 3.

71. Practice. Calculation delta-rule expressions for various activation functions

72. Some problems

The learning rate is important
Too small
Convergence extremely slow
Too large
May not converge
The result may
converge to
a local minimum.
Possible decision:
Using adaptive
learning rate

73. Some problems

Overfitting
The number of hidden neurons is very important, it defines the
complexity of the decision boundary:
Too few
Underfit the data – it does not have enough free
parameters to fit the training data well.
Too many
Overfit the data – NN learns the insignificant details
Try different number and use validation set to choose the
best one.
Start small and increase the number until satisfactory results
are obtained.

74.

What constitutes a “good” training
set?
Samples must represent the general
population
Samples must contain members of each
class
Samples in each class must contain a
wide range of variations or noise effect

75. Practice with multilayer perceptron

1.
2.
Using MLP for noisy digits
recognition &
Using MLP for time series
forecasting.
- Training set preparation.
- MLP learning in Deductor software.
- Estimation the error.

76. Recurrent neural networks

Capable to influence to themselves
by means of recurrences, e.g. by
including the network output in the
following computation steps.
Hopfield neural network
Hamming neural network

77. Hopfield network

1. Invented by John Hopfield in 1982.
2. Content-addressable memory with binary threshold nodes (-1,1 or 0,1)
3. wij=wji, wii=0
yi t 1 F w ji y j t Ti
j 1
j i
Y t 1 F S t ; S t W T Y t T
S S1 ,..., S n Y y1 ,..., yn
T
T T1 ,..., Tn
w11
w
W 21
...
wn1
T
T
w12
w22
...
wn 2
... w1n
... w2 n
wii 0
... ...
... wnn

78. Hopfield network

79. Hopfield network as associative memory

80. Using hopfield network as associative memory

y1 y11
2 2
y
y
Y 1
... ...
L
y L y1
y12
y22
...
y2L
y1n
2
... yn
... ...
L
... yn
...
y i {0;1}
Hebbian rule
W (2Y I )T (2Y I ) I
1 1 1
L 0 0
where I 1 1 1 I 0 L 0
1 1 1
0 0 L

81. Hopfield network as associative memory

1.
2.
Take noisy pattern y
Realize iterations
yi (t 1) F w ji y j (t )
j
1, S 0
sign( S )
0, S 0
3.
Until we will not reach stable state
(attractor)

82. Example

83. Practice with Hopfield network

Realization of the associative memory
based on Hopfield Neural Network
Working with multiple patterns.
Recognition of the original and noisy
patterns.
Investigation of the properties and
constraints of the associative memory
based on Hopfield network.

84. Hamming network

R. Lippman (1987)
Hamming network is two-network bipolar classifier. The first
layer is single-layer perceptron. It calculates hamming distance
between the vectors. The second network is Hopfield network.

85. Hamming network

X 1 x11 ,..., x1n X 2 x12 ,..., xn2 X m x1m ,..., xnm
wij xij / 2 , T j n / 2
yj dj
d j Hamming distance between input pattern and j stored pattern
1, if k j
vkj
, e const ,0 e 1 / m
e, if k j
S j , S j 0
z j F S j
0, S j 0

86. Hamming network working algorithm

Define weights wij, Tj
Get input pattern and initialize
Hopfield weights
Make iterations in Hopfield network
until we get stable output.
Take output neuron with 1 value.

87. Self-organizing maps

88. Self-organizing maps

Unsupervised Training
The training set only consists of input
patterns.
The neural network adjusts its own weights
so that similar inputs cause similar outputs.
The network identifies the patterns and
differences in the inputs without any
external assistance

89. Self-organizing maps (SOM)

A self-organizing map (SOM) is a type of 
artificial neural network that is trained using 
unsupervised learning to
produce
a
lowdimensional
(typically
two-dimensional),
discretized representation of the input space of
the training samples, called a map.
Self-organizing maps are different from other
artificial neural networks in the sense that they
use a neighborhood function to preserve the 
topological properties of the input space.
The model was first described as an artificial
neural network by the Finnish professor Teuvo
Kohonen. 

90. Self-organizing maps

We only ask which neuron is active at the
moment.
We are not interested in the exact output of the
neuron but in knowing which neuron provides
output.
These networks widely used for clustering
SOMs (like our brain) decide the task of
mapping
a
high-dimensional
input
(N
dimensions) onto areas in a low-dimensional
grid of cells (G dimensions).

91.

92. Scheme of training of self-organizing map

93. Competitive learning

Competitive learning is a form of unsupervised
learning in artificial neural networks, in which nodes
compete for the right to respond to a subset of the
input data
S j wij xi W j X T
i
where X x1 ,..., xn input pattern
W j w1 j ,..., wnj
winner take all rule
if S k max S j then
j
1, if j k
y j F S j
0, if j k

94. Competitive learning

Dj X Wj
x1 w1 j 2 ... xn wnj 2
Neuron winner Dk min D j
j
Wk (t 1) Wk (t ) X (t ) Wk (t )

95. Vector quantization

It works by dividing a large set of
points (vectors) into groups having
approximately the same number of
points closest to them. Each group is
represented by its centroid point, as
in k-means and
some
other clustering algorithms.

96. Vector quantization

Choose random weights from
[0;1].
t=1
Take all input patterns Xl,l=1,L
D lj X l W j
pattern
recognition
x1 w1 j 2 ... xn wnj 2
Neuron winner Dkl min D lj
j
Applications:
data compression
Video codecs
wij (t 1) wij (t ) (t ) xi wij (t ) , j k
QuickTime
wij (t 1) wij (t ), j k
(t ) 1 / t
Cinepak
Indeo etc.
t=t+1
Audio codecs
Ogg Vorbis
TwinVQ
DTS etc.

97. Kohonen Maps

98. Kohonen maps

Neightborh ood function for neuron winner
h p,k,t e
u k u p
2
2 2 ( t )
W p (t 1) W p (t ) t h p, k , t X (t ) W p (t ) for winner neuron

99. Kohonen maps learning procedure

1.
Choose random weights from [0;1].
2.
t=1
3.
Take input pattern Xl and calculate Dij=(Xl-Wij),where i,j=1,m
4.
Detect winner neuron D(k1,k2)=min(Dij)
5.
Calculate for every output neuron
uk u p
h p,k,t e
6.
2
2 2 ( t )
Modify weights by following way
W p (t 1) W p (t ) t h p, k , t X (t ) W p (t ) for winner neuron
Repeat steps 3-6 for all input patterns

100. Training and Testing

101. Training

The goal is to achieve a balance
between correct responses for the
training patterns and correct
responses for new patterns.

102. Training and Verification

The set of all known samples is
broken into two independent sets
Training set
A group of samples used to train the neural
network
Testing set
A group of samples used to test the
performance of the neural network
Used to estimate the error rate

103. Verification

Provides an unbiased test of the quality
of the network
Common error is to “test” the neural
network using the same samples that
were used to train the neural network.
The network was optimized on these
samples, and will obviously perform well on
them
Doesn’t give any indication as to how well
the network will be able to classify inputs
that weren’t in the training set

104. Summary (Discussion)

Artificial neural networks are inspired by the learning
processes that take place in biological systems.
Artificial neurons and neural networks try to imitate
the working mechanisms of their biological
counterparts.
Learning can be perceived as an optimisation
process.
Biological neural learning happens by the
modification of the synaptic strength. Artificial neural
networks learn in the same way.
The synapse strength modification rules for artificial
neural networks can be derived by applying
mathematical optimisation methods.

105. Summary

Learning tasks of artificial neural networks can
be reformulated as function approximation
tasks.
Neural networks can be considered as nonlinear
function approximating tools (i.e., linear
combinations of nonlinear basis functions),
where the parameters of the networks should be
found by applying optimisation methods.
The optimisation is done with respect to the
approximation error measure.
In general it is enough to have a single hidden
layer neural network (MLP or other) to learn the
approximation of a nonlinear function.

106. Questions and Comments

107. Thank you for attention

English     Русский