Documente Academic
Documente Profesional
Documente Cultură
NEURON
1
Materi
2
Models of Neuron
Neuron is an information processing unit
A set of synapses or connecting links
characterized by weight or strength
An adder
summing the input signals weighted by synapses
a linear combiner
An activation function
also called squashing function
squash (limits) the output to some finite values
3
Nonlinear model of a neuron (I)
4
Analogy
5
Nonlinear model of a neuron (II)
6
Types of Activation Function
7
Activation Functions...
8
Types of Activation Function
Identity Function
9
Types of Activation Function
Also known as
Heaviside function
10
Activation Functions - sigmoid
11
Activation Functions - sigmoid
12
Activation Function value range
13
Stochastic Model of a Neuron
14
Stochastic Model of a Neuron
15
Decision boundaries
16
E.g. Decision Surface of a
Perceptron
17
Linear Separability
18
Rugby players & Ballet dancers
19
Training the neuron
20
The artificial neuron learning
Supervised Learning
Unsupervised Learning
21
Supervised Learning
22
Unsupervised Learning
(Learning without a teacher)
23
Neuron Learning
24
Training
25
Simple network
26
Learning algorithm
27
Learning algorithm
28
Learning algorithm
29
Training the neuron
30
Training the neuron
31
Learning in Neural Networks
32
Network Architecture
34
Feedforward Networks
35
Feedforward Networks
(Recurrent or dynamic systems)
36
Feedforward Networks
(Recurrent or dynamic systems)
37
Neural network as directed Graph
38
Neural network as directed Graph
depending on w
stable, linear divergence, exponential divergence
we are interested in the case of |w| <1 ; infinite
memory
output depends on inputs of infinite past
NN with feedback loop : recurrent network
40
Neural Processing
Recall
The process of computation of an output o for a
given input x performed by the ANN.
It’s objective is to retrieve the information, i.e., to
decode the stored content which must have been
encoded in the network previously
Autoassociation
A network is presented a pattern similar to a
member of the stored set, autoassociation
associates the input pattern with the closest stored
pattern.
Autoassociation: reconstruction of incomplete or
noisy image
41
Neural Processing
Heteroassociation:
The network associates the input pattern with pairs
of patterns stored
42
Neural Processing
Classification
A set of patterns is already divided into a number
of classes, or categories
When an input pattern is presented, the classifier
recalls the information regarding the class
membership of the input pattern
The classes are expressed by discrete
valued output vectors, thus the output neurons of
the classifier employ binary activation functions
A special case of heteroassociation
43
Neural Processing
Recognition
If the desired response is the class number, but
the input pattern doesn’t exactly corresponding
to any of the patterns in the stored set
44
Neural Processing
Clustering
Unsupervised classification of patterns/objects
without providing information about the actual
classes
The network must discover for itself any existing
patterns, regularities, separating properties, etc.
While discovering these, the network undergoes
change of its parameters, which is called Self-
organization
45
Neural Processing
patterns stored
46
Summary