On completion
of this course, the student should be able to understand neural network
architectures and learning algorithms and also be able to apply neural networks to real
classification problems.
architectures and learning algorithms and also be able to apply neural networks to real
classification problems.
Introduction,:
Humans And Computers. The Structure of the Brain.Pattern Recognition.The
Basic Neuron & Learning Algorithm. The Multilayer Backpropagation Network (BPN). The
Multilayer Perceptron Algorithm.Kohonen Self-Organising Networks.The BAM (Bidirectional
Auto-Associative Memory) Network.The Hopfield Memory Networks.Simulated Annealing.
Mean Field Theory, Spin Glasses, Constraint Satisfaction, The Travelling Salesman Problem,
The Elastic Net. The Counter Propagation Network (CNP). CNP Data Processing.Adoptive
Resonance Memory. The Initialization Phase, The Recognition Phase, The Comparison Phase,
Vigilance Threshold, The Search Phase. ART Algorithm, Training Art Network, Scaling The
Feedforward Weights, The Training Cycle, Classification.
Basic Neuron & Learning Algorithm. The Multilayer Backpropagation Network (BPN). The
Multilayer Perceptron Algorithm.Kohonen Self-Organising Networks.The BAM (Bidirectional
Auto-Associative Memory) Network.The Hopfield Memory Networks.Simulated Annealing.
Mean Field Theory, Spin Glasses, Constraint Satisfaction, The Travelling Salesman Problem,
The Elastic Net. The Counter Propagation Network (CNP). CNP Data Processing.Adoptive
Resonance Memory. The Initialization Phase, The Recognition Phase, The Comparison Phase,
Vigilance Threshold, The Search Phase. ART Algorithm, Training Art Network, Scaling The
Feedforward Weights, The Training Cycle, Classification.
1.
Introduction:
Humans And Computers, The Structure of The Brain, Learning in
Machines, The Differences .[TB1: Ch1]
Machines, The Differences .[TB1: Ch1]
2.
Pattern
Recognition: Feature Vectors And Feature Space, Discriminant Functions,
Classification Techniques, Linear Classifiers.[TB1: Ch2]
Classification Techniques, Linear Classifiers.[TB1: Ch2]
3.
The Basic
Neuron: Modeling the Single Neuron, The Perceptron Learning Algorithm
(Hebbian Learning), Widrow-Hoff Delta Rule, Limitations of Perceptrons, The End of
the Line? [TB1: Ch 3]
(Hebbian Learning), Widrow-Hoff Delta Rule, Limitations of Perceptrons, The End of
the Line? [TB1: Ch 3]
4.
The Multilayer
Backpropagation Network (BPN), The Generalized Delta Rule, Updates
of Output-Layer Weights, Updates of Hidden Layer Weights [TB2: Ch3, TB1: Ch4]
of Output-Layer Weights, Updates of Hidden Layer Weights [TB2: Ch3, TB1: Ch4]
5.
The Multilayer
Perceptron Algorithm(Summary), The XOR Problem Revisited,
Multilayer Perceptron as Classifier, Learning Difficulties [TB1: Ch4, TB2: Ch3]
Multilayer Perceptron as Classifier, Learning Difficulties [TB1: Ch4, TB2: Ch3]
6.
Kohonen
Self-Organising Networks: Introduction, The Self-Organisation Concept, The
Kohonen Algorithm, Weights Training, Initialising The Weights, Reducing The
Neighbourhood [TB1: Ch5]
Kohonen Algorithm, Weights Training, Initialising The Weights, Reducing The
Neighbourhood [TB1: Ch5]
7.
The BAM
(Bidirectional Auto-Associative Memory) Network: Associative-Memory
Definitions, BAM Architecture, BAM Processing, BAM Mathematics, BAM Energy
Function [TB2: Ch4]
Definitions, BAM Architecture, BAM Processing, BAM Mathematics, BAM Energy
Function [TB2: Ch4]
8.
The Hopfield
Memory Networks: Introduction, The Hopfield Network Algorithm, The
Energy Landscape, Storing Patterns, Recall The Stored Patterns [TB1: Ch6]
Energy Landscape, Storing Patterns, Recall The Stored Patterns [TB1: Ch6]
9.
Simulated
Annealing: Statistical-Mechanics Concepts, Real And Simulated Annealing,
The Boltzman Machine, Basic Architecture and Processing Algorithm, Learning in
Boltzman Machines [TB2: Ch5]
The Boltzman Machine, Basic Architecture and Processing Algorithm, Learning in
Boltzman Machines [TB2: Ch5]
10.
Mean Field
Theory, Spin Glasses, Constraint Satisfaction, The Travelling Salesman
Problem, The
Elastic Net [TBI: Ch6]
11.
The Counter
Propagation Network (CNP): CNP Building Blocks, The Input Layer, The
Instar and Its Learning Algorithm, Competitive Networks, The Outstar. [TB2: Ch6]
Instar and Its Learning Algorithm, Competitive Networks, The Outstar. [TB2: Ch6]
12.
CNP Data
Processing, Forward Mapping, Training Algorithm of the CNP , Adoptive
Resonance Memory: Adoptive Resonance Theory (ART), The ART Architecture, ART-1
Operation CNP [TB2: Ch6, TB1: Ch7]
Resonance Memory: Adoptive Resonance Theory (ART), The ART Architecture, ART-1
Operation CNP [TB2: Ch6, TB1: Ch7]
13.
The
Initialization Phase, The Recognition Phase, The Comparison Phase, Vigilance
Threshold, The Search Phase. [TB1: Ch7]
Threshold, The Search Phase. [TB1: Ch7]
14.
ART Algorithm,
Training Art Network, Scaling The Feedforward Weights, The Training
Cycle, Classification. [TB1: Ch7]
Cycle, Classification. [TB1: Ch7]
1.
Neural
Computing (an introduction); by R Beal and T Jackson; Institute of Physics
Publishing, Techno House, Redcliffe Way, Bristol BS1 6NX, UK; (1994).
Publishing, Techno House, Redcliffe Way, Bristol BS1 6NX, UK; (1994).
2.
Neural
Networks (Algorithms, Applications, and Programming Techniques) by James A.
Freeman and David M. Skapura; Published by Pearson Education (Singapure) Pte. Ltd.,
Indian Branch, 482 F.I.E. Patpargannj, Delhi 110 092, India; (2004).
Freeman and David M. Skapura; Published by Pearson Education (Singapure) Pte. Ltd.,
Indian Branch, 482 F.I.E. Patpargannj, Delhi 110 092, India; (2004).
•
Neural
Networks and Pattern Recognition by C. Bishop, Oxford University Press, (1996).
ISBN-10: 0198538642
ISBN-10: 0198538642
r(J
•
Neural
Networks and Learning Machines by Simon O. Haykin, Prentice Hall; 3 Edition
(November 28, 2008). ISBN-10: 0131471392
(November 28, 2008). ISBN-10: 0131471392
Note: This
content is obtained from official documents of University of Sargodha and
applied on BS Computer Science for Main Campus, Sub
Campuses, and Affiliated Colleges.
0 comments:
Post a Comment