Вы находитесь на странице: 1из 5

COURSE DESCRIPTION

1. Study program
1.1 Institution Bucharest Academy of Economic Studies
1.2 Faculty Faculty of Cybernetics, Statistics and Informatics for Economy
1.3 Department Department of Informatics and Cybernetics for Economy
1.4 Field of study Economy
1.5 Study cycle Bachelor
1.6 Study program Informatics for Economy

2. Course
2.1 Course name Evolutionary programming and genetic algorithms
Course code 16.0205IF2.2-0003
2.2 Course holder prof.dr. Cătălina Lucia Cocianu,
conf.dr. Cristian Răzvan Uscatu
2.3 Seminary holder prof.dr. Cătălina Lucia Cocianu,
conf.dr. Cristian Răzvan Uscatu,
lect.dr. Radu Mogoș
2.4 Year of study 2 2.5 Semester 2 2.6 Evaluation type EC 2.7 Course regime O

3.Total estimated time (hours per semester)


3.1 Hours per week of which: 3.3 seminary /
4 2 2
3.2 course laboratory
3.4 Total hours in study plan of which: 3.6 seminary /
56 28 28
3.5 course laboratory
Time allocation ore
Study of the manual, course materials, bibliography and course notes 17
Additional documentation in library, specialty electronic platforms and in the field 7
Preparing for seminars/laboratories, homework, reports, essays and portfolios 16
Tutoring 2
Examination 2
Other activities
3.7 Total hours for individual study 44
3.9 Total hours per semester 100
3.10 Number of credits 4

4.Prerequisites (if any)


4.1 curricula Computer programming basics
Programming algorithms and techniques
Probabilities and mathematical statistics
4.2 competencies

5.Requirements (if any)


5.1. for course Course hall with:
- at least 150 seats,
- video-projector cu the possibility of linking a laptop,
- whiteboard and markers.
5.2. for seminary/laboratory Seminar hall with:
- at least 25 seats,
- at least 20 computers with MatLab and Visual Studio 2010 installed
- white board and marker.
Students must carry am external data storage device (memory stick), with a
reasonable capacity (minimum 2 GB, recommended 4GB), on which to save all
laboratory work for future reference, both for individual study and future
laboratory sessions.
General conditions Full presence and active participation to course/laboratory is recommended.
If a class is missed, it is the student’s duty to determine which chapter was
studied and take measures to recover lost activity. Many themes can only be
mastered through exercise, frequent absences implicitly affecting the final
result.
Attendance to practical tests is mandatory, as there is no make up test.
Use of mobile phones and similar devices during classes is forbidden. Also, it is
forbidden to use class computers for anything else than class activities
(especially not for accessing social platforms of any kind).

6. Specific competencies acquired


 C4 - developing software components by using data structures, algorithms, techniques and
competencies competencies
Transversal Professional

modern programming languages.

7. Course objectives (as result of specific competencies acquired)


7.1 General objective of the In depth study of evolutionary algorithms and their use for solving
course optimization problems in economy.
7.2 Specific objectives - Understanding of fundamental concepts of evolutionary computation.
- Understanding, learning and using genetic algorithms, evolutionary
algorithms and genetic programming.
- Using evolutionary algorithms to solve some classes of optimization
problems with performance criterion (optimization of investment
portfolio etc.).

8. Content
8.1 Course Teaching techniques Notes
1. Introductory lecture: course objectives and targeted Lecture based on multimedia Course support
competencies, methods and instruments used, presentation and resources is given to the
documentation sources, evaluation requirements. and case studies. students at the
Theoretical basis of evolutionary computation. beginning of the
Evolutionary algorithms. Components and classification lecture
– introduction.
2. From direct search to stochastic search. Hill-climbing Idem Idem
and simulated annealing algorithms.
3. Evolutionary algorithms components and classification Idem Idem
– representation, variation operators (recombination
and mutation).
4. Evolutionary algorithms components and classification
–population models, evaluation function, survivor
selection mechanism (replacement).
5. Genetic algorithms. Representation of individuals: Idem Idem
binary representation, floating point representation,
permutation representation.
6. Genetic algorithms. Population models, evaluation Idem Idem
function, mutation operator.
7. Genetic algorithms. The recombination operator.
8. Genetic algorithms. Selection operators. Idem Idem
9. Evolutionary strategies. The general scheme. The Idem Idem
uncorrelated/correlated mutation operators
10. Evolutionary strategies. Parent selection, survivor
selection. Self adaptation
11. Portfolio optimization. Optimal portfolio with n Idem Idem
stocks.
12. Genetic algorithms for portfolio optimization: Idem Idem
representation, population models, evaluation function.
13. Genetic algorithms for portfolio optimization: Idem Idem
mutation and crossover operators, selection operator.
14. Written test. Evaluation Course and
seminar support
Bibliography
1. Electronic course support, available at
2. Cocianu Cătălina, Uscatu Cristian: Programare evolutivă şi algoritmi genetici, Editura ASE, 2015
3. Eiben, A. E., Smith, J. E., Introduction to Evolutionary Computing, Springer-Verlag, 2003
4. Fulcher, J., Jain, L. C. (Eds.), Computational Intelligence: A Compendium,Springer-Verlag, 2008
5. Engelbrecht, A.P., Computational Intelligence. An Introduction, John Wiley & Sons, 2007

8.2 Seminary / Laboratory / Applied work / Stages Teaching techniques Notes


1. Themes to be studied and bibliography. Evaluation Teacher-student Computers with Internet
requirements for seminary activity. introduction. access and adequate
Basic elements of Matlab. Presentation. software.
2. Implementation of hill-climbing algorithm. Computer Idem
implementation and
practical examples
3. Evolutionary computation. Example: knapsack
Idem Idem
problem.
4. Evolutionary computation. Genetic algorithms.
Example: the 8 (n) queens problem.
5. Genetic algorithms. Example: job shop scheduling Idem Idem
problem. Variants for representation and solving.
6. Evolutionary strategies and programming for Idem Idem
optimization problems.
7. Portfolio optimization problems: risk minimization,
return maximization, risk minimization for a given Idem Idem
minimum return level.
8. Practical test.
9. Evolutionary strategies. Applications
10. Evolutionary strategies. A comparative analysis
11. Genetic algorithms for portfolio optimization without
Idem Idem
short selling.
12. Genetic algorithms for portfolio optimization with
Idem Idem
short selling. – 2 seminaries
13. Presentation of a project that includes solving a
particular optimization problem (chose a type of
evolutionary algorithm, population representation, Evaluation Idem
description of operators used, algorithm parameters,
stop condition).
Bibliography
1. Electronic course support, available at
2. Cocianu Cătălina, Uscatu Cristian: Programare evolutivă şi algoritmi genetici, Editura ASE, 2015
3. Eiben, A. E., Smith, J. E., Introduction to Evolutionary Computing, Springer-Verlag, 2003
4. Fulcher, J., Jain, L. C. (Eds.), Computational Intelligence: A Compendium,Springer-Verlag, 2008
5. Engelbrecht, A.P., Computational Intelligence. An Introduction, John Wiley & Sons, 2007

9. Coordination of course content with expectations of epistemic community professional associations and
employers from the field related to the study program
Discussions on the curricula with specialists and employers regarding the need and importance of using
evolutionary algorithms for development and optimization of components for data analysis informatics
systems.
Empowering the future specialists with abilities and skills in using modern techniques and methods that reflect
the current state of the field.
Use of presented evolutionary algorithms to design optimization modules for performance criteria functions.
Use of evolutionary algorithms and hybrid techniques for various classes of applications in industry, economy,
medicine etc.

10.Evaluation
10.3 % of final
Activity 10.1 Criteria 10.2 Evaluation method
grade
10.4 Examination In depth study and learning of
evolutionary computation and genetic
Written exam 50%
algorithms concepts and their use to
solve given optimization problems.
10.5 Seminary / Application of acquired knowledge to
laboratory solve given problems, interpret and Practical test. 20%
argue the results.
Ability to apply acquired knowledge to
elaborate a project (as described in the Presentation of a project 30%
last seminar activity).
10.6 Minimal performance standard
Understanding, learning and using fundamental elements of evolutionary computation to solve optimization
problems (chose a type of evolutionary algorithm, population representation, description of operators used,
algorithm parameters, stop condition).
Although learning is often a social and collaborative process, the grades must reflect individual effort,
knowledge and skills. Solutions with unspecific similarities will receive zero points.
Any attempt to cheat during examinations will result in elimination, with zero points, and further measures
will be taken in accordance with the institution regulations.

Date Course holder signature Seminary holder signature

........................... ............................................. .............................................


Department approval date Department director signature

............................................ ..............................................................

Вам также может понравиться