COURSE PRESENTATION FORM – ACADEMIC YEAR 2011 / 2012

COURSE NAME Machine Learning: Algorithms and Applications

COURSE CODE

72065 (MSc 270) / 70189 (BSc, MSc 509)

LECTURER

Floriano Zini

TEACHING ASSISTANT

--

TEACHING LANGUAGE

English

CREDIT POINTS

4

LECTURE HOURS 24

EXERCISE HOURS

12

OFFICE HOURS

LECTURER

During the lecture time span, Wednesday, 13.45-15:45

Faculty of CS, POS Building, piazza Domenicani 3

, office 2.18

OFFICE HOURS

TEACHING ASSISTANT

--

PREREQUISITES

• Computer programming

• Calculus

• Linear algebra

•

Probability

OBJECTIVES

There are many problems in real life for which we cannot directly

implement a computer program that solves them. Think, for example, to

the recognition of spoken words, to a robot that has to reach a specific

location in an unknown environment, to a user interface that adapts to

the specific biometrics of the user, or to an advisory system that helps

clinicians in finding the correct diagnosis for a patient.

The goal of Machine Learning is the design and implementation of

algorithms that allow computers to automatically learn from data or past

experience how to improve their performance at tasks and problems like

these.

In this introductory course, the students will familiarize with the main

machine learning algorithms, will understand their

strengths and

weaknesses, will learn which techniques are more appropriate for which

problems, and will study how to design a learning experiment and

evaluate the goodness of the learned solution.

Examples of real

applications will be presented.

In the exercise hours, the students will apply the studied machine

learning techniques to solve sample problems in practice.

SYLLABUS • Types and sample application of machine learning

•

Bayesian learning

• Instance based learning

• Genetic algorithms

• Neural networks

• Kernel methods

• Combining multiple learners

• Reinforcement learning

•

Design and analysis of machine learning experiments

TEACHING FORMAT

F

rontal lectures, labs,

and projects.

ASSESSMENT • Project in small teams (max 2 people) [40 % of mark]

• Final exam (written) [60 % or mark]

Project part must be passed before taking the written exam. In case of a

positive mark the project will count for all 3 regular exam sessions.

READING LIST

Textbook

There is not a singe textbook. Suggested books are:

• Ethem ALPAYDIN, Introduction to Machine Learning, 2nd edition,

The MIT Press, 2010

• Tom Mitchell, Machine Learning, 2nd edition, McGraw Hill, 1997.

Suggested readings:

•

Other readings will be suggested during the course.

SOFTWARE USED

• Weka - http://www.cs.waikato.ac.nz/ml/weka/

• NetBeans IDE - http://netbeans.org/

•

Java

LEARNING OUTCOME The students will learn the fundamental principles of machine learning,

the main learning techniques and how to design and evaluate a machine

learning experiment. They will be able to put in practice the acquired

knowledge, by implementing simple learning applications.

COURSE PAGE http://www.inf.unibz.it/~zini/ML

## Comments 0

Log in to post a comment