Download MITM613

MITM 613
Intelligent System
Chapter 0: Introduction
Abdul Rahim Ahmad
2
Contents
 Introduction
 Objectives
 Outcomes
 Chapters
 Plan
 Assessment
 References
 Conclusion and Expectations
Abdul
Rahim
Ahmad
3
Introduction
 This course emphasises on the methods and
techniques that can be used to develop intelligent
systems.
 knowledge-based techniques
 expert and rule-based system
 object-oriented and frame-based systems
 intelligent agents.
 computational intelligence or Machine Learning techniques
 neural networks and its similar tools
 genetic algorithms
 Fuzzy logic
Abdul
Rahim
Ahmad
 a hybrid of both.
4
Objectives
 To provide understanding of intelligent systems
and the various methods and tools in
implementing Intelligent Systems.
 To demonstrate the implementation of individual
methods within the scope of Intelligent systems
 To compare the pros and cons of each method
of developing Intelligent Systems.
 To develop the ability to implement a particular
Intelligent system of choice
Abdul
Rahim
Ahmad
5
Outcomes
At the end of the course, you should be able to:
 Explain the various methods of implementing
Intelligent systems
 Describe the issues involved in each method of
implementing an Intelligent System.
 Describe the tools that can be used.
 Develop a particular intelligent system of choice
in a class project environment.
Abdul
Rahim
Ahmad
6
Main text
 Adrian A. Hopgood, Intelligent Systems for
Engineers and Scientists, 2nd Edition, CRC
Publication (2000).
 http://www.adrianhopgood.com/
Abdul
Rahim
Ahmad
7
Abdul
Rahim
Ahmad
Chapters from Hopgood















Chapter one: Introduction
Includes Fuzzy
Chapter two: Rule-based systems
Logic
Chapter three: Dealing with uncertainty
Chapter four: Object-oriented systems
Specifically on
Chapter five: Intelligent agents
Genetic Algorithm
Chapter six: Symbolic learning
Chapter seven: Optimization algorithms
Chapter eight: Neural networks
Additional
Chapter nine: Hybrid systems
Chapter – Support
Chapter ten: Tools and languages
Vector Machine
Chapter eleven: Systems for interpretation and diagnosis
Chapter twelve: Systems for design and selection
Chapter thirteen: Systems for planning
Chapter fourteen: Systems for control
Chapter fifteen: Concluding remarks
Chapter one: Introduction
1.1 Intelligent systems
1.2 Knowledge-based systems
1.3 The knowledge base
1.4 Deduction, abduction, and induction
1.5 The inference engine
1.6 Declarative and procedural programming
1.7 Expert systems
1.8 Knowledge acquisition
1.9 Search
1.10 Computational intelligence
1.11 Integration with other software
Chapter two: Rule-based systems
2.1 Rules and facts
2.2 A rule-based system for boiler control
2.3 Rule examination and rule firing
2.4 Maintaining consistency
2.5 The closed-world assumption
2.6 Use of variables within rules
2.7 Forward-chaining (a data-driven strategy)
2.7.1 Single and multiple instantiation of variables
2.7.2 Rete algorithm
2.8 Conflict resolution
2.8.1 First come, first served
2.8.2 Priority values
2.8.3 Metarules
2.9 Backward-chaining (a goal-driven strategy)
2.9.1 The backward-chaining mechanism
2.9.2 Implementation of backward-chaining
2.9.3 Variations of backward-chaining
2.10 A hybrid strategy
2.11 Explanation facilities
Chapter three: Dealing with
uncertainty
3.1 Sources of uncertainty
3.2 Bayesian updating
3.3 Certainty theory
3.3.1 Introduction
3.2.1 Representing uncertainty by
probability
3.3.2 Making uncertain hypotheses
3.2.2 Direct application of Bayes’
theorem
3.3.4 A worked example of certainty
theory
3.2.3 Likelihood ratios
3.3.5 Discussion of the worked
example
3.2.4 Using the likelihood ratios
3.2.5 Dealing with uncertain evidence
3.2.6 Combining evidence
3.2.7 Combining Bayesian rules with
production rules
3.2.8 A worked example of Bayesian
updating
3.2.9 Discussion of the worked
example
3.2.10 Advantages and disadvantages
of Bayesian updating
3.3.3 Logical combinations of evidence
3.3.6 Relating certainty factors to
probabilities
3.4 Possibility theory: fuzzy sets and
fuzzy logic
3.4.1 Crisp sets and fuzzy sets
3.4.2 Fuzzy rules
3.4.3 Defuzzification
3.5 Other techniques
3.5.1 Dempster–Shafer theory of
evidence
3.5.2 Inferno
Chapter four: Object-oriented systems
Skipped
Chapter five: Intelligent agents
5.1 Characteristics of an intelligent agent
5.2 Agents and objects
5.3 Agent architectures
5.3.1 Logic-based architectures
5.3.2 Emergent behavior architectures
5.3.3 Knowledge-level architectures
5.3.4 Layered architectures
5.4 Multiagent systems
5.4.1 Benefits of a multiagent system
5.4.2 Building a multiagent system
5.4.3 Communication between agents
Chapter six: Symbolic learning
Skipped
Chapter seven: Optimization
algorithms
7.1 Optimization
7.2 The search space
7.3 Searching the search space
7.4 Hill-climbing and gradient
descent algorithms
7.4.1 Hill-climbing
7.4.2 Steepest gradient descent
or ascent
7.4.3 Gradient-proportional
descent
7.4.4 Conjugate gradient
descent or ascent
7.5 Simulated annealing

7.6 Genetic
algorithms
 7.6.1 The basic GA
 7.6.2 Selection
 7.6.3 Gray code
 7.6.4 Variable length
chromosomes
 7.6.5 Building block
hypothesis
 7.6.6 Selecting GA
parameters
 7.6.7 Monitoring evolution
 7.6.8 Lamarckian inheritance
 7.6.9 Finding multiple optima
 7.6.10 Genetic programming
Chapter eight: Neural networks
8.1 Introduction
8.2 Neural network applications
8.2.1 Nonlinear estimation
8.2.2 Classification
8.2.3 Clustering
8.2.4 Content-addressable memory
8.3 Nodes and interconnections
8.4 Single and multilayer perceptrons
8.4.1 Network topology
8.4.2 Perceptrons as classifiers
8.4.3 Training a perceptron
8.4.4 Hierarchical perceptrons
8.4.5 Some practical considerations
8.5 The Hopfield network
8.6 MAXNET
8.7 The Hamming network
8.8 Adaptive Resonance Theory (ART) networks
8.9 Kohonen self-organizing networks
8.10 Radial basis function networks
Chapter nine: Hybrid systems
9.1 Convergence of techniques
9.2 Blackboard systems
9.3 Genetic-fuzzy systems
9.4 Neuro-fuzzy systems
9.5 Genetic-neural systems
9.6 Clarifying and verifying neural networks
9.7 Learning classifier systems
Chapter ten: Tools and languages
 10.1 A range of intelligent systems tools
 10.2 Expert system shells
 10.3 Toolkits and libraries
 10.4 Artificial intelligence languages
 10.4.1 Lists
 10.4.2 Other data types
 10.4.3 Programming environments
 10.5 Lisp
 10.5.1 Background
 10.5.2 Lisp functions
 10.5.3 A worked example
 10.6 Prolog
 10.6.1 Background
 10.6.2 A worked example
 10.6.3 Backtracking in Prolog
 10.7 Comparison of AI languages
19
Assessment
Abdul
Rahim
Ahmad
 Assignments (3 x 5)
15%
 Projects (best of 2 x 15)
15%
 Mid. Semester Examination
30%
 Final Examination
40%
20
All References
 Adrian A. Hopgood, Intelligent Systems for
Engineers and Scientists, 2nd Edition, CRC
Publication (2000).
 Vojislav Kecman, Learning and Soft Computing:
Support Vector Machines, Neural Networks, and
Fuzzy Logic Models (Complex Adaptive Systems),
MIT Press 2001
 Artificial Intelligence, Elain Rich, Kevin Knight,
Shivashanker Nair, McGraw Hill, 2009
Abdul
Rahim
Ahmad
21
Conclusion/Expectations
 Able to explain fundamental concepts.
 Able to implement selected methods.
 Appreciation for using intelligent methods in
other field.
Abdul
Rahim
Ahmad