Curriculum and course descriptions MMIS program

Timetable

Project for specialization

The project for specialization includes an in-depth study on the different topics, which were educated during the various courses, and to put them into practise. First, the students will do a study on related literature. Then, computational methods are developed and implemented. Finally, a report is written and a presentation is given about the work. Each year a number of different topics will be discussed. The project for specialization is focused on the preparation of the students for their actual graduation work.

Graduation Work

The graduation work includes an in-depth study of related literature, the development and implementation of computational methods, and the writing of the thesis. The topics could be chosen from a list of available items or, with the approval of the graduation mentor, the topic could be defined by the student. The size of the graduation work corresponds to the time which is required to obtain a thorough study on the subject, to provide a solid implementation, a good positioning of external traineeships (if applicable), and to create the possibility to follow additional courses e.g. the student is allowed to follow a course on which the graduation work is based on.

Constrained/Free choice

A distinction is made between a 'constrainted choice' and 'free choice'. The 'constrainted choice' should be chosen from a list of courses. The 'free choice' is a course which might be given by another institute or university.

Course descriptions MMIS programma

Title: Multimedia Information Retrieval

Lecturer Th. Gevers

Goals Understanding of multimedia information for retrieval purposes

Contents With the growth of the Internet and developments in imaging technology,
very large digital image and video archives have been created and used
in numerous applications. Together with the increase in the number of
pictorial archives, demands are also growing for methodologies
and techniques to store and retrieve pictorial entities from
large image archives.

In this course a broad range of techniques are studied to access
multimedia information including multimedia features (synonyms for text,
color and shape invariants for images), multimedia information space
modeling (logic model, vector space model, statistical model),
indexing (kd-trees, inverted file), learning and classification
(nearest neighbor, neural network), user interaction (active learning),
visualization and presentation techniques.

Literature Reader.

Form Lectures and lab sessions.

Title: Multimedia

Lecturer A. W. M. Smeulders

Goals Understanding of structure, components, and performance of systems for multimedia retrieval

Contents In this course, the following topics are studied:
- Multimedia: 1 definition, 2 application examples, 3 information types, 4 components
- Multimedia analysis: sensor integration
- Multimedia knowledge discovery: speech recognition, text mining, picture learning
- Multimedia systems and standards
- Multimedia multi-modal interaction and visualization
- Semantic access to information
- Personalized information delivery
- Video as the ultimate multimedia
- Multimedia information systems
Literature To be announced.

Form Lectures and lab sessions.

Title: Advanced Database Techniques

Lecturer M.L. Kersten

Goals Understanding of extensions to the relational model, query processing, storage structures and the DBMS architecture to support advanced applications

Contents In this course, we study developments to enhance traditional relational database technology to provide data management for non-administrative applications.
Topics included are XML and Xquery for web-based applications;
temporal models to support applications involving time; spatial extensions to support e.g. Geographical information systems; association rule discover over databases to support data mining; and support for querying streaming data such as they appear in sensor networks.

Literature Reader based on web-accessible literature.

Form Lectures and lab sessions.

Title: Machine Learning: Pattern Recognition

Lecturer B. Kröse

Goals Understanding of novel methods and algorithms for statistical pattern
Recognition

Required
Background Basic programming skills, statistics, and linear algebra

Contents - Intro: what are the basic issues in Statistical Pattern Recognition?
- Density Estimation, Expectation Maximization
- Linear discriminant analysis
- Kernel methods, support vector machine
- System performance, combining classifiers
- Feature selection, Principal Component Analysis, Multidimensional Scaling
- Clustering, Vector quantization, Self-organizing map

Literature Statistical Pattern Recognition, 2nd Edition, Andrew Webb, ISBN: 0-470-84514-7, Paper, 534 Pages, August 2002.

Form Lectures and lab sessions

Title: Multi-agent systems and distributed AI

Lecturer N. Vlassis

Goals Understanding of principles and algorithms of agent interactions.
Contents The concept of an "intelligent agent" is fundamental in Artificial Intelligence, meaning anything that can perceive the environment and act upon it in order to achieve some goal. In "distributed AI" the focus is no longer on single agents but on "multiagent systems": these are collections of intelligent agents that can interact with each other. Allowing interaction between agents offers new solutions to AI problems but also raises challenging issues: how can a group of agents come up with joint plans? how can they share knowledge? how can they coordinate their actions? In this course we will study some basic principles of agent interactions and how they can be turned into practical algorithms. Particular topics that will be covered include acting under uncertainty, strategic games, common knowledge, coordination, and multiagent learning.

Literature Syllabus and articles.

Form Lectures and lab sessions.
.

Title: Design and organization of autonomous systems

Lecturer F. Groen

Goals Understanding of design and organization of autonomous systems.

Contents This course will focus on the integration aspects when dealing with autonomous systems. Topics which will be presented are: architectures for autonomous systems, models for hiearchical decomposition, internal representations and models, sensor data fusion, reactive behaviour, coolaborating agents.

Literature To be announced.

Form Lectures and an elaborate laboratory assignment in which students work in small groups on a specific integration project.