2. Content
Introduction
History
Capabilities
Applications
Real world implementation
Issues
Advantage/Disadvantage
Conclusions
References
3. Introduction
Brain : A highly complex, non-linear &
parallel computing
Structural constituent : ‘Neurons’
A computer program designed to model
the human brain.
5. Conti…
Our brains are made up of about 10
billion tiny units called neurons .
Tree like Nerve fibres are called
dendrites .
Signals coming into the neuron are
received via junctions called synapses .
6. Artificial Neuron
Dendrites
Synapses
The
information
transmission
happens at
Soma (cell body)
the synapses.
Axon
7. Conti…
A network of interconnected functional
elements.
A NN is trained to recognize and
generalize the relationship between a
set of inputs and outputs.
9. History
Inspiration for development came from
attempts to model the human central
nervous system.
McCulloch-Pitts 1943 Introduced a
simple NN model using electrical
circuits.
10. Conti…
Hebb 1949 Wrote that neural pathways
are strengthened every time they are
used.
Minsky 1954: Learning Machine.
Rosenblatt’s Perceptron 1957.
11. Conti…
Widrow 1960 : Adaline model.
Recent work includes Boltzmann
machines, competitive learning models,
multilayer networks, and adaptive
resonance theory models.
12. Artificial neurons
Neurons work by processing
information.
x1
x2 w1
n Output
x3 w2 z = ∑ wi xi ; y = H ( z )
i =1 y
w3
… ..
. w
n-1
xn-1
wn
xn
The McCulloch-Pitts model
15. Real World Implementation
Logical reasoning
Pattern recognition
Planning
Genetic programming
Common sense knowledge
Representation
Control system
16. Issues
Complex programs
Difficult to implement
Machine prediction may not be accurate
Human beings may lose their importance
17. Advantage
Pattern recognition
Does not need to be reprogrammed
Implemented in any application
Adaptive learning
Self-Organisation
Real Time Operation
Fault Tolerance