Dr. G. Y.

Pathrikar College of Computer Science & Information

Technology, Aurangabad

Teaching Evaluation Scheme

Course code Course Title Scheme Internal External Credit
Total
L T P CA MSE TW ESE PR
DSIT-501 Basics of Neural 3 - - 20 20 - 60 - 3
Networks

L- Lecture, T-Tutorial, P-Practical, CA- Continuous Assessment, MSE- Mid Semester Examination, ESE-
End Semester Examination, PR-Practical, TW-Term Work.
Prerequisite : Basic knowledge of linear algebra, calculus and logic.

Course To study learning and modeling of the algorithms of Neural Networks.

Objective:

Course Outcome: After completion of the course the student will be able to:

Cognitive Level
Sr.No. Course Outcome as per Bloom
Taxonomy

1. Understand the modeling and applications of Neural Networks L1, L2

Algorithms used in Neural Networks L1, L2,

2.
L3,L4,L5

Detailed Syllabus:
 Dr. G. Y. Pathrikar College of Computer Science & Information
Technology, Aurangabad

Cognitive
No. Level of
Section Topics to be covered of attainment as
Lect. per
Taxonomy
UNIT I 15
INTRODUCTION: Feedforward Neural Networks:
Artificial Neurons, Neural Networks and
Architectures: Neuron Abstraction, Neuron Signal
Functions, Mathematical Preliminaries, Neural Networks
Defined, Architectures: Feed forward and Feedback, Salient
A Properties and Application Domains of Neural Network L1 , L2, L3
Geometry of Binary Threshold Neurons and Their
Network: Patterns Recognition and Data Classification,
Convex Sets, Convex Hulls and Linear Separability, Space
of Boolean Functions, Binary Neurons are pattern
apacity
of a simple Threshold Logic Neuron, Revisiting the XOR
Problem, Multilayer Networks.
UNIT II 15
SUPERVISED LEARNING: Supervised Learning I:
Perceptrons and LMS: Learning and Memory, From Synapses
to Behaviour: The Case of Aplysia, Learning Algorithms, Error
Correction and Gradient Descent Rules, The Learning Objective
B for TLNs, Pattern space and Weight Space, Perceptron Learning L1, L2, L3,
Algorithm, Perceptron Convergence Theorem, Perceptron L4, L5

Surface and its Geometry, Steepest Descent Search with Exact

Application of LMS to Noise Cancellation.

UNIT-III 15
Supervised Learning II: Backpropagation and Beyond:
Multilayered Network Architectures, Backpropagation Learning L1, L2, L3,
C
Algorithm, Structure Growing Algorithms, Fast Relatives of L4, L5
Backpropagation, Universal Function Approximation and Neural
Networks, Applications of Feedforward Neural
Networks,Reinforcement Learning
 Dr. G. Y. Pathrikar College of Computer Science & Information
Technology, Aurangabad

Books and References:

SR.NO. Title Author Publishers Edition

1. Satish Kumar Tata McGraw Hill

2. Introduction to neural networks using Sivanandam, S TATA

MATLAB 6.0 by, Sumathi, S N
Deepa McGraw HILL

3 Neural networks A comprehensive Simon Hhaykin Pearson Education 2nd

foundations. edition
2004
4. Artificial neural networks B.Yegnanarayana Prentice Hall of
India P Ltd 2005.

Strategies to help student to gain their attentions & Presenting Content of Course:

Ask Question Presentation Brainstorming

Show Video Interactive Lecture Case-Study

Learning Activities Think-pair-share Project Based Learning

Scenario Demonstration Group Discussion

Share a Problem Fish Bowl Flipped Classroom