SOFT COMPUTING

– AN INTRODUCTION

Session - 1
Contents
 Introductionto Computing
 Hard computing
 Soft computing
 Hard Vs Soft computing
 Soft computing Techniques
 Fuzzy Logic
 Neural networks
 Genetic algorithms
 Evolutionary algorithms
 Introduction to computing
Computation :- There will be certain input and
then there is a procedure by which the input can be
converted into some output.

In the context of computing, the input is called the

antecedent and output is called the consequence
and then computing is basically mapping.
So, in other words, computing is nothing
but is a mapping function, mapping from
set of input to output.

Now, this mapping is also alternatively

called as formal method also it is called an
algorithm, so basically algorithm to solve a
problem
Characteristics of Computing
Should provide Precise Solution.

Control action should be unambiguous &

accurate.

 Good for those problem that can be

mathematically modelled.
Hard Computing
 Hard computing is a traditional approach used in computing which
needs an accurately stated analytical model, proposed by Dr. Lotfi
Zadeh. It works on the mathematical model of problem-solving.

 Hard computing is suitable for mathematical problems, although it

might be used to solve real-world problems, it consumes a large
amount of computation time and cost.

 Applications of hard computing are numerical problems, searcing

and sorting techniques, computational geometry etc.

 If we want to solve the deterministic problems, we can use a hard

computing approach. As the problem grows in size and complexity,
the design search space also increases.
Soft Computing
 The term Soft Computing was also coined by inventor of fuzzy logic
Lofti A. Zadeh

 Soft computing deals with approximate models and gives solutions to

complex real-world problems.

 Soft computing is an approach that gives an impression on the human

mind to reason and learns in an environment of uncertainty and
impression.

 Soft computing deals with imprecision, uncertainty, partial truth and

approximation to achieve close resemblance with human decisions.

 The cost of the solution also very low in this kind of computing, it forms
the basis of a considerable amount of machine learning techniques.
Advantages of Soft Computing
 Models based on human reasoning.

 Closer to human thinking and biologically inspired.

 Models can be Linguistic Comprehensible Fast when

computing Effective in practice.
Applications of Soft Computing
 Handwriting Recognition

 Image Processing and Data Compression

 Automotive Systems and Manufacturing

 Soft Computing to Architecture

 Decision-support Systems
Evolutionary problem
Fuzzy logic
Hard Computing Vs Soft Computing
 Hard computing is very accurate and certain whereas the
soft computing model is imprecision tolerant and works on
partial truth and approximation.

 Hard computing is based on a crisp system and binary

logic and soft computing are based on fuzzy logic and
probabilistic reasoning.

 Hard computing works on exact input data. On the contrary

soft computing works on noisy and ambiguous data.

 Hard computing performs sequential computations while

soft computing performs parallel computations on the data.
Hard Computing Vs Soft Computing
 Hard computing provides accurate results. On the other hand, soft
computing provides approximate results.

 The nature of the hard computing approach is deterministic whereas

soft computing is probabilistic.

 Hard computing uses two-valued logic, on the other hand, soft

computing uses multi-valued logic.

 Hard computing requires programs to be written but soft computing

will emerge its programs.
Techniques in soft Computing
Fuzzy Logic
Neutral Networks
Genetic algorithms
Probabilistic Reasoning
Approximation and Intelligence
◦ Decision making with soft computing(hybrid
technique)
Fuzzy Systems
 Fuzzy system is a superset of conventional logic
that has been extended to handle the concept of
partial truth- truth values between "completely true"
and "completely false".

 Based on fuzzy set theory and fuzzy logic.

 Most applications in control and decision making.

Fuzzy Logic
A Fuzzy control system is a control system based on
Fuzzy logic — a mathematical system that analyzes
analog input values in terms of logical variables that take
on continuous values between 0 and 1, in contrast to
classical or digital logic, which operates on discrete values
of either 1 or 0.
Fuzzy Control
 Fuzzy Logic is a technique to embody human like
thinking into a control system.

A fuzzy controller is designed to emulate human

deductive thinking, that is, the process people use to
infer conclusions from what they know.

 Traditional control approach requires formal modeling

of the physical reality.
Fuzzy Control
 Fuzzy control, which directly uses fuzzy rules is the most
important application in fuzzy theory.

 Using a procedure originated by Ebrahim Mamdani in the

late 70s, three steps are taken to create a fuzzy controlled
machine:
◦ Fuzzification (Using membership functions to graphically describe a
situation)
◦ Rule evaluation (Application of fuzzy rules) 
◦ Defuzzification (Obtaining the crisp or actual results) 
Fuzzy Controller Architecture
Fuzzification
 Fuzzification is the process of making a crisp quantity
fuzzy.

 They carry considerable uncertainty.

 Ifthe form of uncertainty arises because of imprecision or

fuzziness, it can be represented by a membership function.

 Institution method is used for fuzzification of the input

variables.
Defuzzification
 Defuzzification is the conversion of a fuzzy quantity to a
precise quantity.

 Defuzzification techniques :
◦ Max - Membership Principle known as height method is limited to
peaked output junctions given by,

◦ Centroid Method also called center of area, center of gravity given by,
Defuzzification
 Defuzzification techniques :
◦ Weighted Average Method is a valid for symmetrical O/P membership
function given by,

where ∑ denotes an algebraic sum.

◦ Means-Max Membership (middle of maxima) : The MAX membership

can be a plateau rather than a single point given by,
Fuzzy Logic Control System
– Automated Accident Prevention System
 Obstacle Sensor Unit
◦ Sensing Distance

 The sensing distance depends upon the speed of the car. speed
can be controlled by gradual anti skid braking system.

 An automated accident prevention system is necessary to

prevent accidents.

 The fuzzy logic control system can relieve the driver from
tension & prevents accidents.

 This fuzzy control unit results in an accident-free world.

Fuzzy Logic Control System
– Automated Accident Prevention System
 Input Membership Function for Velocity:

 Input Membership Function for Distance:

Fuzzy Logic Control System
– Automated Accident Prevention System
 Output Membership Function For Brake system:
 Neural Networks
 Neural Network is an information processing paradigm
that is inspired by the way biological nervous systems,
such as the brain, process information.

 Essentially a function approximator.

 Transforms inputs into outputs to the best of its ability.

Genetic Algorithm
A genetic algorithm (or GA) is a search technique used
in computing to find true or approximate solutions to
optimization and search problems.

 GAs are categorized as global search heuristics.

 GAs are a particular class of evolutionary algorithms

that use techniques inspired by evolutionary biology
such as inheritance, mutation, selection, and crossover
(also called recombination).
Basic Genetic Algorithm
 Start with a large “population” of randomly generated
“attempted solutions” to a problem

 Repeatedly do the following:

◦ – Evaluate each of the attempted solutions
◦ – (probabilistically) keep a subset of the best solutions
◦ – Use these solutions to generate a new population

 Quit when you have a satisfactory solution (or you run

out of time)
Evolutionary Algorithms
 An optimization technique based on ideas of adaptation and
evolution. It belongs to the general class of evolutionary computation
or artificial evolution methodologies.

 EA are stochastic search and optimization heuristics derived from the

classic evolution theory, which are implemented on computers in
most cases.

 If only those individuals of a population reproduce, which meet a

certain selection criteria, and the other individuals of the population
die, the population will converge to those individuals that best meet
the selection criteria.

 Population dynamics follow the basic rule of Darwin evolution

theory, which can be described in short as the “survival of the fittest.”
 Classification of EA methods

•  In a "genetic algorithm," the problem is encoded in a

series of bit strings that are manipulated by the algorithm;

• in an "evolutionary algorithm," the decision variables

and problem functions are used directly.