EVOLUTION OF INTELLIGENT/
SOFT COMPUTING
A Presentation by
[Link]
Associate Professor
Dept. of Electrical Engg
NIT Arunachal Pradesh
Yupia – 791112
1
INTELLIGENCE & AUTONOMY
Increasingly capable controllers: a
worthy challenge necessitating Value
increased ability along various Judgment
dimensions of intelligence
Sensor World Behavior
Processing Model Generation
Sensors Structure Actuators
World
2
INTELLIGENT SYSTEMS
Intelligence: System must perform
meaningful operations.
Interprets information.
Comprehends the relations between the
phenomena or objects.
Applies the acquired information to new
conditions.
3
OBJECTIVES OF INTELLIGENT SYSTEMS
• Everyday routines of human beings: vision,
language processing, common sense
reasoning, learning, robotics.
• Artificial routines: games, mathematics,
logic, programming.
• Expert systems (ES)
4
TRADITIONAL APPROACHES
• Mathematical models:
Black boxes, number
crunching.
• Rule-based systems (crisp
& bivalent):
Large rule bases.
5
WHY & WHAT do we COMPUTE
Real World
Problem
Mathematical Mathematical
Model Model
PROBLEM
SOLUTION SOLVING
METHODS
6
HARD COMPUTING
• Conventional/Traditional method of
problem solving.
• Does things exactly as per formula/theory.
• Inspired by Boolean logic
• Sufficient and suitable for ancient systems.
(1800 & Early Nienties).
• Can represent only few values.
7
HARD COMPUTING
• Processes usually are Repetitive:
Change in every iteration only affects
the values but not the process.
• Requires exact and accurate data.
• Imprecision and uncertainty are
undesirable.
• Difficult to represent vagueness and
real world situations.
8
Machine Learning
• Machine Learning (ML) is constructing computer
programs that develop solutions and improve with
experience
• Solves problems which can not be solved by enumerative
methods or calculus-based techniques
• Intuition is to model human way of solving some
problems which require experience
• When the relationships between all system variables is
completely understood ML is not needed
9
A Generic ML System
x1 y1
x2 y2
System
xN h1 , h2 , ..., h K
yM
Input Variables: x x1 , x2 ,..., xN
Hidden Variables: h h1 , h2 ,..., hK
Output Variables: y y1 , y2 ,..., yK
10
Machine Intelligence: A core concept for grouping various advanced
technologies with Pattern Recognition and Learning
Hybrid Systems
Knowledge-based Neuro-fuzzy
Genetic neural
Systems
Fuzzy genetic
Probabilistic reasoning Fuzzy neuro
Approximate reasoning genetic Data Driven
Case based reasoning Systems
Neural network
system
Machine
Evolutionary
Intelligence computing
Fuzzy logic
Non-linear
Rough sets
Dynamics
Chaos theory
Pattern recognition Rescaled range
and learning analysis (wavelet)
Fractal analysis
11
SOFT COMPUTING
• Recent Method
• Optimize and approximate things
the natural human way
• Tolerance for Imprecision and
uncertainty is exploited
• Can represent a range of values
• Inspired by fuzzy logic
• Evolutionary process : change in
every iteration not only affects
the values but the process also
12
Generating a Model
Cleaning Noisy Data, Missing
values
Stochastic Hill Climbing
13
INTELLIGENT CONTROL (IC)
Objective:
Mimic human (linguistic) reasoning
Main constituents:
- Fuzzy systems
- Neural networks
- Evolutionary computing
- Probabilistic reasoning
14
CONSTITUENTS OF INTELLIGENT
CONTROLLERS
• Fuzzy systems => imprecision
• Neural networks => learning
• Probabilistic reasoning => uncertainty
• Evolutionary computing => optimization
Over 24 000 publications today
15
FRAMEWORK OF INTELLIGENT
CONTROL COMPUTING
16
IC: A USER-FRIENDLY INTERFACE
S o f t c o m p u tin g
approach
L in g u is tic w o r ld
S o f t d a ta
M a th e m a tic a l w o r ld I n te r p r e ta tio n s
H a r d d a ta U n d e r s ta n d in g
Q u a n tita tiv e m e th o d s E x p la n a tio n s
B iv a le n t r e a s o n in g Q u a lita tiv e m e th o d s
B iv a le n t o r m u ltiv a le n t
r e a s o n in g
P h e n o m e n o n u n d e r s tu d y
17
Intelligent Controllers are present everywhere.
All we need is an eye for them.
18
We are witnessing a revolution in the
making as scientists from all different
disciplines discover that complexity has a
strict architecture. We have come to grasp
the important knowledge of intelligent
controllers.
19
IC TODAY (ZADEH)
• Computing with words (CW)
• Theory of information granulation (TFIG)
• Computational theory of perceptions (CTP)
20
POSSIBLE IC DATA & OPERATIONS
• Numeric data:
5, about 5, 5 to 6, about 5 to 6
• Linguistic data:
cheap, very big, not high, medium or bad
• Functions & relations:
f(x), about f(x), fairly similar, much greater
21
NEURAL NETWORKS (NN, 1940'S)
• Neural networks offer
a powerful method to
explore, classify, and
identify patterns in
data.
Neurons
Inputs Outputs
(1 layer)
• Neuron: y=wixi
22
INTERCONNECTIONS IN BRAIN
23
MACHINE LEARNING (SUPERVISED)
• Pattern recognition
Peach based on training
data.
Instructor • Classification
supervised by
instructor.
Plum • Neural (crisp or
?
fuzzy), neuro-fuzzy
and fuzzy models.
24
MACHINE LEARNING (UNSUPERVISED)
• Pattern recognition
Peach
based on training
Nectarine
data.
• Classification based
on structure of data
Plum (clustering).
• Neural (crisp or
fuzzy), neuro-fuzzy
Labeling and fuzzy models.
25
MACHINE LEARNING (UNSUPERVISED)
• Self-organized maps
Peach
(Kohonen).
Nectarine • Fuzzy c-means
(Bezdek).
• Subclust (Yager, Chiu).
Plum
Labeling
26
NEURAL NETWORK ARCHITECTURES
27
OPERATION OF A NEURAL NET
x0 w0j
- Bias
x1 w1j
f output y
xn wnj
Input weight weighted Activation
vector x vector w sum function
28
FUZZY SYSTEMS (ZADEH, 1960'S)
• Deal with imprecise entities in automated environments
(computer environments)
• Base on fuzzy set theory and fuzzy logic.
• Most applications in control and decision making
Omron’s fuzzy processor
29
FUZZY LOGIC CONTROLLERS (FLC)
30
(NEURO)-FUZZY SYSTEM CONSTRUCTION
Training Fuzzy rules Experts
data (SOM, c-means
etc.)
Control System Tuning
data evaluation (NN)
(errors)
New system
31
ADVANTAGES OF INTELLIGENT COMPUTING
AND CONTROL
• Models based on human reasoning.
• Models can be
- linguistic
- simple (no number crunching),
- comprehensible (no black boxes),
- fast when computing,
- good in practice.
32
INTELLIGENT
CONTROLLER EXAMPLES
33
INTELLIGENT CONTROLLER APPLICATIONS
• Heavy industry
(Matsushita, Siemens,
Stora-Enso)
• Home appliances
(Canon, Sony, Goldstar,
Siemens)
• Automobiles (Nissan,
Mitsubishi, Daimler-
Chrysler, BMW,
Volkswagen)
• Spacecrafts (NASA)
34
INTELLIGENT CONTROLLER APPLICATIONS
•supplier evaluation for •hospital stay prediction,
sample testing, •TV commercial slot evaluation,
•customer targeting, •address matching,
•sequencing, •fuzzy cluster analysis,
•scheduling, •sales prognosis for mail order
•optimizing R&D house,
•projects, •multi-criteria optimization etc.
•knowledge-based •(source: FuzzyTech)
prognosis,
•fuzzy data analysis
35
INTELLIGENT CONTROLLER APPLICATIONS
• Fuzzy scoring for mortgage applicants,
• creditworthiness assessment,
• fuzzy-enhanced score card for lease risk assessment,
• risk profile analysis,
• insurance fraud detection,
• cash supply optimization,
• foreign exchange trading,
• insider
• trading surveillance,
• investor classification etc.
• Source: FuzzyTech
36
INTELLIGENT CONTROLLER APPLICATIONS:
ROBOTICS
Helpmate Robotic
Courier
Entertainment
Robot
37
INTELLIGENT CONTROLLER APPLICATIONS:
OTHERS
•Statistics
•Social sciences
•Behavioural sciences
•Biology
•Medicine
38
IC AND FUTURE
IC and conventional methods should be used
in combination.
39
REFERENCES
•J. Bezdek & S. Pal, Fuzzy models for pattern recognition (IEEE Press,
New York, 1992).
•L. Zadeh, Fuzzy logic = Computing with words, IEEE Transactions on
Fuzzy Systems, vol. 2, pp. 103-111, 1996.
•L. Zadeh, From Computing with Numbers to Computing with Words --
From Manipulation of Measurements to Manipulation of Perceptions,
IEEE Transactions on Circuits and Systems, 45, 1999, 105-119.
•L. Zadeh, Toward a theory of fuzzy information granulation and its
centrality in human reasoning and fuzzy logic, Fuzzy Sets and Systems
90/2 (1997) 111-127.
•H.-J. Zimmermann, Fuzzy set theory and its applications (Kluwer,
Dordrecht, 1991).
•[Link]: Evolutionary Computation: A New Transactions, IEEE Trans.
On Evolutionary Computation, 1-1, 1(1998)
•[Link] eta. Ed.:Fuzzy Control, Wiley (2000)
•[Link] eta.: Soft-Computing as a Breakthrough, Vol.39, Mach 2000, J.
of SICE (2000) (in Japanese)
40
THANK YOU
41
