SOFT

COMPUTING
What is Soft Computing?
Soft computing is an emerging approach to computing
which parallel the remarkable ability of the human mind to
reason and learn in a environment of uncertainty and
imprecision.
Some of it’s principle components includes:
Neural Network(NN)
Fuzzy Logic(FL)
Genetic Algorithm(GA)
These methodologies form the core of soft computing.
GOALS OF SOFT COMPUTING
The main goal of soft computing is to develop
intelligent machines to provide solutions to real
world problems, which are not modeled, or too
difficult to model mathematically.
It’s aim is to exploit the tolerance for
Approximation, Uncertainty, Imprecision, and
Partial Truth in order to achieve close
resemblance with human like decision making.
Hard Computing and Soft Computing
SOFT COMPUTING -
DEVELOPMENT HISTORY
Soft = Evolutionary + Neural + Fuzzy
Computing Computing Network Logic
Zadeh Rechenberg McCulloch Zadeh
1981 1960 1943 1965

Evolutionary = Genetic + Evolution + Evolutionary + Genetic

Computing Programming Strategies programming Algorithms
Rechenberg Koza Rechenberg Fogel Holland
1960 1992 1965 1962 1970
NEURAL NETWORKS
An NN, in general, is a highly interconnected network of a
large number of processing elements called neurons in an
architecture inspired by the brain.
NN Characteristics are:-
Mapping Capabilities / Pattern Association
Generalisation
Robustness
Fault Tolerance
Parallel and High speed information processing
 Neuron
Biological neuron

Model of a neuron
ANN ARCHITECTURES
X X
y1 z1
1 1
y1

X
y2 X
2 z2
2

X y2
y3 X
3 z3
3
Input Layer Output Layer Input Layer Hidden Layer
1.Single Layer Feedforward Output Layer
Network 2.Multilayer Feedforward Network

X
Xi - Input Neuron
z1
1
y1

X
Yi - Hidden /Output
z2
2
y2 Neuron
X z3
3
Zi - Output Neuron
Input Layer Hidden Layer
LEARNING METHODS OF ANN
NN Learning
algorithms

SSupervis
Unsupervised Reinforced
ed
Learning Learning
Learning

Error
Correcti Stochasti Competitiv
c Hebbian
on e

Least
Mean Backpropagati
Square on
FUZZY LOGIC
Fuzzy set theory proposed in 1965 by A. Zadeh is a
generalization of classical set theory.
In classical set theory, an element either belong to
or does not belong to a set and hence, such set
are termed as crisp set. But in fuzzy set, many
degrees of membership (between o/1) are allowed
FUZZY VERSES CRISP
FUZZY CRISP
IS R AM HONEST ?
IS WATER COLORLESS ?
Extremely
Honest(1) YES!(1)

Very
FUZZY Honest(0.8) CRISP

Honest at
Times(0.4) NO!(0)

Extremely
Dishonest(
0)
OPERTIONS
CRISP FUZZY

1.Union 1.Union
2.Intersection 2.Intersection
3.Complement
3.Complement
4.Equality
4.Difference
5.Difference
6.Disjunctive Sum
PROPERTIES
CRISP FUZZY
 Commutativity  Commutativity
 Associativity  Associativity
 Distributivity  Distributivity
 Idempotence  Idempotence
 Identity  Identity
 Law Of Absorption  Law Of Absorption
 Transitivity  Transitivity
 Involution  Involution
 De Morgan’s Law  De Morgan’s Law
 Law Of the Excluded
Middle
 Law Of Contradiction
GENETIC ALGORITHM
Genetic Algorithms initiated and developed in the
early 1970’s by John Holland are unorthodox
search and optimization algorithms, which mimic
some of the process of natural evolution.
Gas perform directed random search through a given
set of alternative with the aim of finding the best
alternative with respect tp the given criteria of
goodness.
These criteria are required to be expressed in terms
of an object function which is usually referred to as
What is a Genetic Algorithm?

• Simple Idea: A Genetic Algorithm (GA) is a way computers

can solve problems by mimicking how living things evolve
over time.
• Inspiration: Just like nature selects the best animals to
survive and reproduce, GA picks the best solutions to a
problem and combines them to create better ones.
• Why Use It? It helps solve tough problems where other
methods might struggle.
How Does Genetic Algorithm Work?

1.Start with a Group: Think of a bunch of guesses (solutions)

to a problem.
2.Check the Guesses: See how good each guess is (this is
called "fitness").
3.Pick the Best: Keep the best guesses and let them
"reproduce" (combine them to make new guesses).
4.Add Some Randomness: Make small random changes
(mutations) to the new guesses.
5.Repeat: Keep doing this until you find a great solution!
Why is Genetic Algorithm Useful?

• Learns Over Time: Just like animals get better over

generations, GA improves guesses step by step.
• Handles Tough Problems: Great for complex problems like
finding the best route, making schedules, or even helping
machines learn.
• Doesn’t Need Exact Rules: Works even when we don’t know
all the details or exact formulas to solve the problem.
Where Can We Use Genetic Algorithms?

• Real-Life Examples:
• Travel Planning: Finding the best route for a trip.
• Game Strategies: Helping computers learn the best way to
play games.
• Designing Stuff: Creating efficient designs in engineering or
architecture.
• Machine Learning: Helping machines learn by selecting the
best features or settings.
BENEFITS OF GENETIC ALGORITHM
Easy to understand.
We always get an answer and the answer gets
better with time.
Good for noisy environment.
Flexible in forming building blocks for hybrid
application.
Has substantial history and range of use.
Supports multi-objective optimization.
Modular, separate from application.
APPLICATION OF SOFT COMPUTING
Consumer appliance like AC, Refrigerators,
Heaters, Washing machine.
Robotics like Emotional Pet robots.
Food preparation appliances like Rice cookers
and Microwave.
Game playing like Poker, checker etc.
Thanks..