20IS603 Architecture of Intelligent Systems
Credits: 3-0-0-3
THIS PAGE contains information about the subject 20IS603 Architecture of Intelligent Systems offered in the second semester of academic year 2020 - 2021 for M.Tech. Industrial Intelligent Systems students. It tells you everything about this course, including its aims, syllabus, and operation.
· What is this course for?
· What you will learn
· Lecture schedule
· Assignments
· Supplementary materials
· References
· What is this course for?
· What you will learn
· Lecture schedule
· Assignments
· Supplementary materials
· References
What is this course for?
Artificial Intelligence has acquired widespread acceptance and approval among enterprises, business applications and research communities. Several startups in recent times have endorsed the discipline. Artificial Intelligence is a broad field that promises to simulate numerous innate human skills like Automatic Programming, Case-Based Reasoning, Neural Networks, Fuzzy Logic, Decision-Making, Expert Systems, Natural Language Processing, Pattern Recognition and Speech Recognition etc. A.I. technologies bring more complex data-analysis features to existing applications. cognitive systems can perceive their environment and react accordingly in an intelligent manner. Another significant feature of these intelligent systems is their ability to learn and adapt to complex and changing environments, requirements and users. Such systems are especially relevant in the areas of robotics, virtual agents, multimedia and web information systems.
What you will learn?
Intelligent systems is a broad term, covering a range of computing techniques that have emerged from research into artificial intelligence. It includes symbolic approaches— in which knowledge is explicitly expressed in words and symbols—and numerical approaches. This course is an essential introduction to a wide range of intelligent systems techniques. You will learn about knowledge-based systems, computational intelligence, and hybrid systems. A comprehensive overview of common practice in designing intelligent systems will be discussed. After taking this class, students should be able to describe the computing technique required to design an intelligent system.
Course Outcomes
CO1: Understand the characteristics of knowledge base systems (BTL – 2)
CO2: Apply the object-oriented concepts in intelligent systems (BTL – 3)
CO3: Identify the characteristics and architectures of multi agent systems (BTL – 2)
CO4: Implement different algorithms for multi-agent systems. (BTL – 3)
Artificial Intelligence has acquired widespread acceptance and approval among enterprises, business applications and research communities. Several startups in recent times have endorsed the discipline. Artificial Intelligence is a broad field that promises to simulate numerous innate human skills like Automatic Programming, Case-Based Reasoning, Neural Networks, Fuzzy Logic, Decision-Making, Expert Systems, Natural Language Processing, Pattern Recognition and Speech Recognition etc. A.I. technologies bring more complex data-analysis features to existing applications. cognitive systems can perceive their environment and react accordingly in an intelligent manner. Another significant feature of these intelligent systems is their ability to learn and adapt to complex and changing environments, requirements and users. Such systems are especially relevant in the areas of robotics, virtual agents, multimedia and web information systems.
What you will learn?
Intelligent systems is a broad term, covering a range of computing techniques that have emerged from research into artificial intelligence. It includes symbolic approaches— in which knowledge is explicitly expressed in words and symbols—and numerical approaches. This course is an essential introduction to a wide range of intelligent systems techniques. You will learn about knowledge-based systems, computational intelligence, and hybrid systems. A comprehensive overview of common practice in designing intelligent systems will be discussed. After taking this class, students should be able to describe the computing technique required to design an intelligent system.
Course Outcomes
CO1: Understand the characteristics of knowledge base systems (BTL – 2)
CO2: Apply the object-oriented concepts in intelligent systems (BTL – 3)
CO3: Identify the characteristics and architectures of multi agent systems (BTL – 2)
CO4: Implement different algorithms for multi-agent systems. (BTL – 3)
Evaluation
Method of evaluation is by Continuous Assessment and an End-of-Semester examination.
Continuous Assessment (Theory)- 50%
Periodical 1 & 2 - 30%
Quiz / Assignment - 10%
Term paper - 10%
End-of-Semester Examination - 50%
If you do not turn in an assignment on time or miss an exam, you will receive a grade of zero, unless you have prearranged approval by your faculty in-charge. There will be no rescheduling of exams or quizzes. Exams and quizzes are excused only for faculty approved medical reasons.
Continuous Assessment (Theory)- 50%
Periodical 1 & 2 - 30%
Quiz / Assignment - 10%
Term paper - 10%
End-of-Semester Examination - 50%
If you do not turn in an assignment on time or miss an exam, you will receive a grade of zero, unless you have prearranged approval by your faculty in-charge. There will be no rescheduling of exams or quizzes. Exams and quizzes are excused only for faculty approved medical reasons.
Suggested Readings
1. Adrian A. Hopgood, “Intelligent systems for engineers and scientists”, Second Edition, CRC press, 2001
2. Crina Grosan, Ajith Abraham, “Intelligent Systems: A Modern Approach “,Springer-Verlag, 2011
3. Bogdan M. Wilamowski, J. David Irwin, “The Industrial Electronics Handbook. Second Edition: Intelligent Systems”, CRC Press, 2011
4. Abraham-Kandel, Gideon-Langholz, “Hybrid-Architectures for Intelligent Systems”, CRC-Press, 1992
1. Adrian A. Hopgood, “Intelligent systems for engineers and scientists”, Second Edition, CRC press, 2001
2. Crina Grosan, Ajith Abraham, “Intelligent Systems: A Modern Approach “,Springer-Verlag, 2011
3. Bogdan M. Wilamowski, J. David Irwin, “The Industrial Electronics Handbook. Second Edition: Intelligent Systems”, CRC Press, 2011
4. Abraham-Kandel, Gideon-Langholz, “Hybrid-Architectures for Intelligent Systems”, CRC-Press, 1992
Lecture Schedule
_ Note: Lecture Schedule subject to change.
Week
1
2 3 4 5 6 7 8 9 10 11 12 |
Topic
Introduction to the course
Knowledge-based Systems Rule-based Expert systems Rule-based systems Handling Uncertainty Possibility theory Intelligent agents Object-Oriented Systems Symbolic Learning Optimization Algorithms Hybrid Systems Systems for Interpretation and Diagnosis |
Keywords
Overview of intelligent systems, examples and applications, intelligent architectures, importance of course, course objectives, course outcomes, Evaluation pattern
components of knowledge based systems, knowledge base, knowledge representation, organization, manipulation and acquisition Expert systems, development of expert systems, characteristics, Rules and facts, inference networks, reasoning, rule-based expert systems Forward-chaining, Conflict resolution, Backward chaining Sources of uncertainty, types of errors contributing to uncertainty, classical probability theory, conditional probability, Bayesian updating, Certainty theory. Periodical - I Crisp Set theory, Fuzzy logic, Fuzzy sets, Fuzzy operations, Membership functions, defuzzification Characteristics of an intelligent agent, Agent Architectures, Multiagent Systems Data abstraction, Inheritance, Encapsulation, Unified Modeling Language (UML), Dynamic (or late) binding Periodical - II Learning by induction, Techniques for Generalization and Specialization, Case-based reasoning Hill-climbing and gradient descent algorithms, Simulated annealing, Genetic algorithms Convergence of Techniques, Genetic–Neural Systems, Genetic–Fuzzy Systems Deduction and Abduction for Diagnosis, Model-Based Reasoning |
Handouts
|
Reading
Adrian A. Hopgood, “Intelligent systems for engineers & scientists”, Second Edition, CRC press, 2001 |