Special Courses 2 (Artificial Intelligence)

CSM221: ARTIFICIAL INTELLIGENCE: METHODS & MODELS

Lecturers:

Victor N. Kaftasjev,. Associate Prof, Ph.D.. TatianaN. Cheboksarova,. Associate Prof., Ph.D..

Areas and issues in artificial intelligence, including problem-solving by searching, inference, knowledge representation, learning and expert systems, implementation and application of programming languages (LISP) and expert systems' shells.

1. Introduction and the state of the art

2. Problem solving by searching

3. Informed search methods

4. Knowledge and reasoning

5. Inference in first-order logic

6. Building a knowledge bases

7. Logical reasoning systems

8. Planning

9. Expert systems

10. Uncertain knowledge and reasoning

11. Pattern recognition and speech synthesis

12. Leaning

1. Ginsberg M- Essentials of Artificial Intelligence. Morgan Kaufmann, San Mateo, California, 1993.

2. Norvige P., Russel S. Artificial Intelligence: A Modem Approach. Prentice Hall, 1995.

CSM222: ARTIFICIAL INTELLIGENCE: ALGORITHMS & PROGRAMMING

Lecturer Vladislav B. Valkovsky, Associate Prof, Ph.D..

The primary aim of the course is to provide the well-established algorithmic techniques in the field ofAl. The implementation language of the algorithms is Prolog. The techniques included in the course cover general areas such as search, backward and forward-chaining methods, rule-based systems, truth maintenance, constraint satisfaction, uncertainty management, randomized algorithms and specific application domains such as temporal reasoning, probabilistic logic reasoning, probabilistic abductive reasoning.

1. On Prolog.

2. Review of graph search techniques.

3. Depth-first, breadth-first, best-first, game-tree search.

4. Backward and forward-chaining.

5. Production systems.

6. Reason, consistency and assumption-based maintenance.

7. Constraint satisfaction.

8. Consistency enforcing.

9. Representing uncertainty in the database.

10. Informal heuristics.

11. Certainty factors.

12. Planning and temporal reasoning.

13. Probabilistic logic reasoning.

14. Probabilistic abductive reasoning.

15. An introduction to randomized algorithms.

16. Randomized equality testing.

17. Selection, sorting and searching by randomized algorithms.

1. Stuart Russell and Peter Norvig. Artificial Intelligence: A Modem Approach. Prentice Hall, 1995.

2. Y.Shoham. Artificial Intelligence Techniques in Prolog. Morgan Kaufmann Publishers Inc, 1994.

3. Rajeev Motwani, Prabhakar Raghavan. Randomized Algorithms. Cambridge University Press, 1995.

CSM223: HARDWARE FOR ARTIFICIAL INTELLIGENCE

Lecturer Michael Kuprijanov, Prof., Dr. Sc.

Organization of hardware tools for intelligent systems support. Functional expanders, specialized coprocessors, hardware platforms. Forward-looking elements for artificial intelligence are analyzed in details. Possibilities of the use ofASIC-s, semicustom VLSI-s and programmable logic are considered.

1. Principles of development of hardware support for intelligent systems.

2. Functional assistants and coprocessors

3. Specialized processors

4. Associative processors.

5. Neural processors and neural networks

6. LISP- and PROLOG- machines.

7. Platforms of leading firms, oriented to realization of elements of artificial intelligence

8. Design ofASTC and CPLD for artificial intelligence

9. Material chosen by instructor

1. Lager and Stubblefield. Artificial Intelligence, 1993.

2. Tanimoto, Elements ofAl, 1994.

CSM224: NEURAL NETWORKS

Lecturer Nikita E. Barabanov, Prof., D.Sc.

Associative memory, capacity, energy function, optimization, discrete and continuous networks, stochastic and deterministic networks, transfer functions, network architectures, leaning with teacher and unsupervised leaning, the most popular kinds of neural networks, applications.

1. Hofield models

2. Stochastic networks

3. Simple perseptron

4. Back propagation

5. Unidirectional, multilayercd networks

6. Network architectures

7. Recurrent networks

8. Learning time sequences

9. Unsupervised leaning,

10. Principal component analysis

11. Learning with competition

12. Kohonen networks

13. Hybrid schemes

1. Introduction to the teory of neural computation// John Henry, Anders Krogh, Richard 0 Palmer. A lecture notes, vol.1., 1991.

2. Practical Neural Networks Recipes in C++// Timothy Masters, academic Press, 1993.

3. Hecht-Nelson Robert. Neurocomputing. Addison-Wesley Publishing Co., Reading, M.A. 1991.

CSM225: EXPERT SYSTEMS

Lecturer Vladimir E. Baltashevich, Associate Prof., Ph.D.

Approaches and methods employed in expert system design and development analysis of selected expert systems. The architecture of expert systems. Knowledge acquisition methodology. The inference engine. The explanatory interface. System building project possible.

1. Knowledge-Based Systems - Introduction. (Definition. Characteristics, evaluation. Generic Categories for knowledge Systems Application).

3. Knowledge representation in expert systems. (Using rules. Using semantic nets. Using frames).

1. A.J. Gonzalez & D.D. Dankel, The Engineering of Knowledge-Based Systems. Prentice Hall, 1993,

2. D.A. Waterman, A Guide to Expert Systems, Addison-Wesley Publishing Company, Reading, MA, 1986.

3. Expert Systems: Principles and Case Studies. Edited by Richard Forsyth. Chapman and Hall, London,1984.

4. F.Hayes-Roth, D.A. Waterman and D.B.Lenat, Building Expert Systems. Addison-Wesley, 1983.