Laurent DUMAS

Muhammad Zaid DAUHOO


Archives : course 2009 , 2010 , 2011, 2012

Course Objectives :

Many problems occuring in industry consist in minimizing (or maximizing) a certain cost function. This course is aimed to present various optimization methods in order to solve such problems.
After a general introduction on  numerical optimization, various optimization methods will be presented:  derivative free optimization, descent-type methods (gradient, newton, BFGS), evolutionary algorithms. A numerical implementation with Scilab of these differents methods will be done during the computer sessions.

Teachers : L . Dumas (Université de Versailles), M.Z. Dauhoo (University of Mauritius)

Course prerequisites :

No specific prerequisites are needed for this course.
It is accessible with basic tools in analysis (functions of n variables).

Syllabus (see timetable ): 

Mercredi 06 février 2013 (14-17h, LD): introduction (lecture1.ppt), derivative free optimization (MDS.sci, NelderMead.sci)

Vendredi 08 février (8h-11h, MZD): steepest descent method, Newton method

Mercredi 20 février (14-17h, LD): linesearch strategy, practical session with MATLAB/SCILAB (énoncé, steepest.sci )

Vendredi 22 février (8h-11h, MZD): linear programming

Mercredi 27 février (14-17h, LD): exercises+computer session : linear programming (simplex.sci)

Vendredi 1 mars (8h-11h, MZD): constrained optimization : theory and algorithms

Vendredi 8 mars (8h-11h,MZD): constrained optimization : theory and algorithms

Vendredi 15 mars (8h-11h, MZD): constrained optimization : theory and algoithms

Mercredi 20 mars (14-17h, LD): simulated annealing (SA.sci), genetic algorithm (GA.sci, GA-binary2011.sci), PSO (ECP2011-PSO.sci)


Examen le  Vendredi 5 avril (énoncé)



Optimisation continue : cours et exercices, J.F. Bonnans, Dunod, 2006.
Numercial optimization, theoretical and practical aspects
: JF Bonnans, JC Gilbert, C. Lemaréchal, C. Sagastizbal, Springer Verlag 2003.
Genetic Algorithms on search, optimization and machine learning : D. Goldberg, 1989

Articles (online):
An introdution to algorithms for non linear optimization : N. Gould, S. Leyffer

Training :

Examen 2012

Examen 2011, rattrapage 2011

Examen 2010 , rattrapage 2010

 Examen 2009 , rattrapage 2009