OVERVIEW

This teaching unit is structured into two modules: Operations Research (OR) and Machine Learning.
OR module introduces models and methods from Operational Research to provide students with tools to address decision-making problems. Students will learn linear programming techniques (including modeling, the simplex algorithm, sensitivity analysis, and duality theory) and integer programming. The module covers graph theory and network flow models.

Machine learning is establishing as a very interesting discipline thanks to the availability of more and more powerful computers and of algorithms allowing its application to the most diverse sectors, including key sectors for economic and societal development. This teaching unit presents the main machine learning methodologies aimed to “pattern recognition”, in particular for the classification of data from signals and images.

AIMS AND CONTENT

LEARNING OUTCOMES

This teaching unit aims to provide students with knowledge of Operations Research models and methods and of Machine Learning methods applied to data, signal and image recognition. In the first case, the aim is to equip students with the tools to address decision-making problems. Specifically, students will learn linear programming techniques (including modeling, the simplex algorithm, sensitivity analysis, and duality theory) as well as integer programming (branch and bound, cutting planes). They will also learn to model decision-making problems in the telecommunications field using graph theory and network flow models. In the Machine Learning area, students will learn to characterize the distribution of a data set and reduce its dimensionality. They will be able to use various classification techniques, with or without training samples, optimizing their structure and parameters and evaluating their performance. They will be able to apply the learned methods to data sets of various types, including signals and images.

PREREQUISITES

Operations Research: Basic knowledge of mathematical analysis, geometry, and computer science.

Machine Learning: Calculus (functions of one or more variables, integrals, functional optimization); probability theory and random variables; matrix calculus.

TEACHERS AND EXAM BOARD

EXAMS

EXAM DESCRIPTION

The exam for the Operations Research module consists of a written test.

The exam of Machine Learning consists of a written test and an oral part. The written text allows students to obtain a maximu mark of 24 out of 30.  The oral part is not mandatory and can be taken by students to obtain a higher mark, up to 30 out of 30, with honors.

ASSESSMENT METHODS

The exam for the Operations Research module is based on a test that requires the students to solve exercises requiring the application of algorithms to the problem classes presented in the course, to answer to short theoretical questions, and to formulate simple combinatorial decision-making problems.

The written test of Machine Learning consists of multiple choice questions and open-ended questions related to the methods presented at lesson; the solution of very simple problems is required, too. A deeper discussion of the methods and the solution of more complex problems is required to pass the oral part of the exam.