Servizi per la didattica

PORTALE DELLA DIDATTICA

01OUWQW, 01OUWNG, 01OUWOQ, 01OUWOV

A.A. 2019/20

Course Language

English

Course degree

Master of science-level of the Bologna process in Mechatronic Engineering - Torino

Master of science-level of the Bologna process in Mathematical Engineering - Torino

Master of science-level of the Bologna process in Electronic Engineering - Torino

Master of science-level of the Bologna process in Computer Engineering - Torino

Course structure

Teaching | Hours |
---|---|

Lezioni | 40 |

Esercitazioni in aula | 20 |

Tutoraggio | 40 |

Teachers

Teacher | Status | SSD | h.Les | h.Ex | h.Lab | h.Tut | Years teaching |
---|---|---|---|---|---|---|---|

Calafiore Giuseppe Carlo | Professore Ordinario | ING-INF/04 | 40 | 20 | 0 | 0 | 8 |

Teaching assistant

Context

SSD | CFU | Activities | Area context |
---|---|---|---|

ING-INF/04 | 6 | D - A scelta dello studente | A scelta dello studente |

2019/20

Optimization is a technology that can be used for devising effective decisions or predictions in a variety of contexts, ranging from production planning, to engineering design, finance, machine learning and data science, to mention just a few. In simplified terms, the process for reaching a decision starts with a phase of construction of a suitable mathematical model for a concrete problem, followed by a phase where the model is solved by means of suitable numerical algorithms. An optimization model typically requires the specification of a quantitative objective criterion of goodness for our decision, which we wish to maximize (or, alternatively, a criterion of cost, which we wish to minimize), as well as the specification of constraints, representing the physical limits of our decision actions, budgets on resources, design requirements than need be met, etc. An optimal design is one who gives the best possible objective value, while satisfying all problem constraints.
While a generic optimization approach typically leads to computational problems that are very hard to solve in practice, convex problems, which are the core of this course, possess special properties that make them amenable to very efficient numerical solution techniques. This course therefore concentrates on recognizing, modeling and solving convex optimization problems that arise in several scientific contexts (engineering, finance, control, data science, etc.), and in showing the relevance of these methodologies in practical applications.

Optimization is a technology that can be used for devising effective decisions or predictions in a variety of contexts, ranging from production planning, to engineering design, finance, machine learning and data science, to mention just a few. In simplified terms, the process for reaching a decision starts with a phase of construction of a suitable mathematical model for a concrete problem, followed by a phase where the model is solved by means of suitable numerical algorithms. An optimization model typically requires the specification of a quantitative objective criterion of goodness for our decision, which we wish to maximize (or, alternatively, a criterion of cost, which we wish to minimize), as well as the specification of constraints, representing the physical limits of our decision actions, budgets on resources, design requirements than need be met, etc. An optimal design is one who gives the best possible objective value, while satisfying all problem constraints.
While a generic optimization approach typically leads to computational problems that are very hard to solve in practice, convex problems, which are the core of this course, possess special properties that make them amenable to very efficient numerical solution techniques. This course therefore concentrates on recognizing, modeling and solving convex optimization problems that arise in several scientific contexts (engineering, finance, control, data science, etc.), and in showing the relevance of these methodologies in practical applications.

To give students the tools and training to recognize convex optimization problems that arise in engineering and other branches of science; to present the basic theory of such problems, concentrating on results that are useful in computation; to give students an understanding of how such problems are solved, and some experience in solving them.

To give students the tools and training to recognize convex optimization problems that arise in engineering and other branches of science; to present the basic theory of such problems, concentrating on results that are useful in computation; to give students an understanding of how such problems are solved, and some experience in solving them.

Good knowledge of linear algebra, geometry, calculus, and some exposure to probability. Exposure to numerical computing, optimization, systems and control theory, and application fields is helpful but not strictly required.

Good knowledge of linear algebra, geometry, calculus, and some exposure to probability. Exposure to numerical computing, optimization, systems and control theory, and application fields is helpful but not strictly required.

Introduction, convex sets and convex functions. Optimization problems in standard form, optimality criteria.
Systems of linear equations, Least Squares (LS), Linear Programming (LP), Ell-one norm optimization, Chebychev approximation.
Application examples: generation of force/torque via thrusters, uniform illumination of patch surfaces, etc.
Quadratic Programming (QP) and Second Order Cone Programming (SOCP).
Application examples (e.g. FIR filter design, antenna array design, sidelobe level minimization, etc.).
Linear Matrix Inequalities (LMI) and semidefinite programming (SDP).
Geometric programming (GP).
Software tools (CVX and/or YALMIP).
Applications: data-fitting, approximation and estimation, uncertain and robust Least Squares, portfolio optimization.
Geometrical problems: containment of poyhedra, classification, Lowner-John ellipsoids, linear discrimination, support vector machines.
Focus topics (mutually exclusive, to be offered alternatively over years):
- LMIs in systems and control theory.
- Sparse optimization and compressed sensing.
- Convex optimization in Finance.
- Convex optimization in algebraic geometry, global polynominal positivity (positivstellensaatz).
- Network optimization, distributed optimization.

Introduction, convex sets and convex functions. Optimization problems in standard form, optimality criteria.
Systems of linear equations, Least Squares (LS), Linear Programming (LP), Ell-one norm optimization, Chebychev approximation.
Application examples: generation of force/torque via thrusters, uniform illumination of patch surfaces, etc.
Quadratic Programming (QP) and Second Order Cone Programming (SOCP).
Application examples (e.g. FIR filter design, antenna array design, sidelobe level minimization, etc.).
Linear Matrix Inequalities (LMI) and semidefinite programming (SDP).
Geometric programming (GP).
Software tools (CVX and/or YALMIP).
Applications: data-fitting, approximation and estimation, uncertain and robust Least Squares, portfolio optimization.
Geometrical problems: containment of poyhedra, classification, Lowner-John ellipsoids, linear discrimination, support vector machines.
Focus topics (mutually exclusive, to be offered alternatively over years):
- LMIs in systems and control theory.
- Sparse optimization and compressed sensing.
- Convex optimization in Finance.
- Convex optimization in algebraic geometry, global polynominal positivity (positivstellensaatz).
- Network optimization, distributed optimization.

The course is organized in a series of lectures (about 1/3 of the course) and computer lab exercises and practice sessions (about 2/3 of the course).

The course is organized in a series of lectures (about 1/3 of the course) and computer lab exercises and practice sessions (about 2/3 of the course).

G.C. Calafiore and L. El Ghaoui, Optimization Models, Cambridge Univ. Press, 2014.
S. Boyd and L. Vandenberghe; Convex Optimization, Cambridge Univ. Press, 2004.

G.C. Calafiore and L. El Ghaoui, Optimization Models, Cambridge Univ. Press, 2014.
S. Boyd and L. Vandenberghe; Convex Optimization, Cambridge Univ. Press, 2004.

The final exam consists in a written test, which will contain a mixture of methodological questions and numerical exercises (to be executed with pen and paper; use of a calculator is allowed) and which will take the form of multiple-choice questionar. The final score will be computed as a weighted average of the written exam score (70%) and a score counting the presence and participation to the lab sessions (30%).

The final exam consists in a written test, which will contain a mixture of methodological questions and numerical exercises (to be executed with pen and paper; use of a calculator is allowed) and which will take the form of multiple-choice questionar. The final score will be computed as a weighted average of the written exam score (70%) and a score counting the presence and participation to the lab sessions (30%).

© Politecnico di Torino

Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY

Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY