Scribd
Upload
55 views25 pages

Complex Pattern Recognition: 6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 1

This document discusses the structure and operation of neural networks. It describes how neural networks are composed of interconnected processing units that are organized based on the problem being solved. The network can be trained to encode information about the problem by adjusting the connections between units. Once trained, the network operates in a production mode where it can classify new input patterns based on what it learned during training. The document also discusses concepts like McCulloch-Pitts networks and Hebbian learning which were influential in the early development of neural network theory.

Uploaded by

NAGAVARSHINI
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
55 views25 pages

Complex Pattern Recognition: 6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 1

This document discusses the structure and operation of neural networks. It describes how neural networks are composed of interconnected processing units that are organized based on the problem being solved. The network can be trained to encode information about the problem by adjusting the connections between units. Once trained, the network operates in a production mode where it can classify new input patterns based on what it learned during training. The document also discusses concepts like McCulloch-Pitts networks and Hebbian learning which were influential in the early development of neural network theory.

Uploaded by

NAGAVARSHINI
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 25

Complex Pattern Recognition

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 1


6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 2
Neural network Structure
Collection of parallel processor connected
together in the form of directed graph
organized such that network structure
lends itself into the problem being
considered.

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 3


Schematic Representation
6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 4
6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 5
Network Operation
• Training and Production Mode
• Training: encoding information about the
problem to be solved
• Network can be modeled in the host
computer as a set of data structures that
represents current state of network

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 6


.. Continue

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 7


Elementary Neurophysiology

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 8


..Continue

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 9


Development of Resting Potential

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 10


Potentials at various locations

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 11


..Continue

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 12


Neurotransmission-Synaptic Jn

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 13


Neural Circuits and Computation

Cell bodies (including Dendrites) represented by large circles


and Small Circles appear at the end of axons

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 14


6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 15
McCulloch-Pitts Theory

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 16


..Continue

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 17


Examples of McCulloch-Pitts
networks

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 18


6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 19
Hebbian Learning

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 20


ANS- PEs

6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 21


6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 22
6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 23
6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 24
6/29/2010 BM1003/Unit1/Lect2/ NNA/ NSR 25

You might also like