Neural Networks: Course Overview

Lecture 01: Course Overview
NEURAL NETWORKS
Md. Mijanur Rahman, Prof. Dr.
Dept. of Computer Science and Engineering, Jatiya Kabi Kazi Nazrul Islam University, Bangladesh.
Email: mijan@jkkniu.edu.bd
INTRODUCTION TO THE COURSE
1. Course Title: Neural Networks
2. Course Code: CSE-5504
3. Course Contents:
 Classifier: linear discriminate function, activation function, Network Architecture:
Fedforard netwrok, single layer, multi layer network, radial function, hopfield network,
Algorithms and methods: supervise and unsupervised learning, issues in learning: data,
training, test, bias, variance and stop training, stopping criteria, complexity, first order
methods of learning, classical supervised learning algorithm, back propagation learning
algorithm: cascade correlation, RBF network, alternative learning: simulated annealing,
genetic algorithms, error correction learning, ensemble based classifier.
COURSE OVERVIEW 2
COURSE OVERVIEW
1. Artificial Neural Network is a branch of Artificial Intelligence
concerned with simulating neurons (cells in the brain responsible
for learning) and applying them to perform learning tasks and
representing knowledge.
 Neurons are information messengers. They use electrical impulses and chemical
signals to transmit information between different areas of the brain, and between the
brain and the rest of the nervous system. The average that we have so far is a total of 86
billion neurons.
2. Through learning, machines are able to perform complex tasks not

achievable (or too complex to be implemented) using conventional
techniques.
COURSE OVERVIEW 3
LEARNING OBJECTIVE
The objective of this course is to provide students with a sound and
comprehensive understanding of artificial neural networks and
machine learning, including subjects of -
1. McCulloch-Pitts model, Hopfield network, Feed-forward and feed-
back network structures, Activation functions;
2. Supervised and unsupervised machine learning algorithms,
Perceptron learning algorithms, Recurrent neural networks,
Weighted networks;
3. Pattern recognition and classification;
4. Dynamic Programming, and deep-learning techniques.
COURSE OVERVIEW 4
LEARNING OBJECTIVE
Course Objectives include-
Objective 1: To provide an introduction to the field of artificial neural
networks and machine learning;
Objective 2: To teach students how to solve practical problems via
implementation of these techniques;
Objective 3: To promote further independent learning on the topics of
artificial neural networks and machine learning;
COURSE OVERVIEW 5
HOW NEURAL NETWORKS USED IN REAL LIFE?
As modern computer become ever more powerful, They are good for Pattern Recognition,
scientists continue to be challenged to use machines Classification and Optimization. This includes
effectively for tasks that are relatively simple for handwriting recognition, face recognition, speech
recognition, text translation, credit card fraud
humans. detection, medical diagnosis and solutions for huge
amounts of data.
Neural networks are a series of algorithms that mimic
the operations of an animal brain to recognize
relationships between vast amounts of data.
Based on examples:
 We learn easily to recognize the letter A or
distinguish a cat from a dog.
 Another common human activity is trying to achieve
a goal that involves maximizing a resource while
satisfying certain constraints.
COURSE OVERVIEW 6
WHY NEURAL NETWORKS?
The development of artificial neural networks began approximately 50 years
ago, motivated by a desire to try both to understand the brain and to emulate
some of it’s strengths.
Interests in neural networks can be attributed to several factors:
 Training techniques have been developed for the more sophisticated network
architectures that arc able to overcome the shortcomings of the early, simple neural
nets.
 High-speed digital computers make the simulation of neural processes more feasible.
 Technology is now available to produce specialized hardware for neural networks.
 Fresh approaches to parallel computing may benefit from the study of biological neural
systems. which are highly parallel.
COURSE OVERVIEW 7
Neural nets are of interest to researchers in many areas for different reasons:
 Electrical engineers find numerous applications in signal processing and control theory.
 Computer engineers are intrigued by the potential for hardware to implement neural nets
efficiently and by applications of neural nets to robotics.
 Computer scientists find that neural nets show promise for difficult problems in areas such
as artificial intelligence and pattern recognition.
 For applied mathematicians, neural nets are a powerful tool for modeling problems for
which the explicit form of the relationships among certain variables is not known.
COURSE OVERVIEW 8
There are various points of view as to the nature of a neural net:
 For example, is it a specialized piece of computer hardware (say, a VLSI chip) or a
computer program?
 The view that neural nets are basically mathematical models of information
processing. They provide a method of representing relationships that is quite
different from Turing machines or computers with stored programs.
 As with other numerical methods, the availability of computer re-sources, either
software or hardware, greatly enhances the usefulness of the approach, especially
for large problems.
The characteristics of biological neural networks that serve as the

inspiration for artificial neural networks, known as neurocomputing.
COURSE OVERVIEW 9
1. They are extremely powerful computational devices
2. Massive parallelism makes them very efficient
3. They can learn and generalize from training data – so there is no need for
enormous feats of programming
4. They are particularly fault tolerant
5. They are very noise tolerant – so they can cope with situations where normal
symbolic systems would have difficulty
6. In principle, they can do anything a symbolic/logic system can do, and more
COURSE OVERVIEW 10
REFERENCE BOOKS
Fundamentals of Neural Networks:
Architectures, Algorithms and Applications
Neural Networks,
By Laurene Fausett By Phil Picton
A Brief Introduction Introduction to

to Neural Networks Neural Networks
By David Kriesel and Deep Learning
By S Raschka
COURSE OVERVIEW 11
LECTURE 01
COURSE OVERVIEW
THE END

Neural Networks: Course Overview

Uploaded by

Copyright:

Available Formats

Neural Networks: Course Overview

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Neural Networks: Course Overview

Uploaded by

Copyright:

Available Formats

Lecture 01: Course Overview

2. Through learning, machines are able to perform complex tasks not

The characteristics of biological neural networks that serve as the

A Brief Introduction Introduction to

You might also like