Welcome to Modelling, Verification, Performance Analysis, Refinement, Synthesis and Machine Learning for Cyber Physical Systems

Organizer: Kim Guldstrand Larsen

Lecturers: Kim Guldstrand Larsen, Marius Mikucionis, Peter Gjøl Jensen, Ulrik Nyman

ECTS: 3

Date/Time: October 2024

Deadline: September 2024

Max no. Of participants: 30

Description: Timed automata and games, priced timed automata and energy automata have emerged as useful formalisms for modeling real-time and energy-aware systems as found in several embedded and cyber-physical systems. During the last 20 years the real-time model checker UPPAAL has been developed allowing for efficient verification of hard timing constraints of timed automata. Moreover a number of significant branches exists, e.g. UPPAAL CORA providing efficient support for optimization, and UPPAAL TIGA allowing for automatic synthesis of strategies for given safety and liveness objectives, and ECDAR supports refinement and compositional development of real-time systems. Also the branch UPPAAL SMC, provides a highly scalable new engine supporting (distributed) statistical model checking of stochastic hybrid automata, and most recently the new branch UPPAAL STRATEGO supporting safe and optimal strategies for stochastic hybrid games by combining symbolic methods with machine learning. The course will review the various branches of UPPAAL, the corresponding modeling formalisms as well as the symbolic or statistical algorithms applied. Also, examples on applications of the tools suite will be given to a range of real-time and cyber-physical examples including schedulability and performance evaluation of mixed criticality systems, modeling and analysis of biological systems, energy-aware wireless sensor networks, smart grids and smart houses and battery scheduling.

Prerequisites: A general background in computer science or a related engineering discipline with basic knowledge about finite-state automata, probability theory and ordinary differential equations. Some basic programming skills are needed.

Important information concerning PhD courses: We have over some time experienced problems with no-show for both project and general courses. It has now reached a point where we are forced to take action. Therefore, the Doctoral School has decided to introduce a no-show fee of DKK 3000 for each course where the student does not show up. Cancellations are accepted no later than 2 weeks before start of the course. Registered illness is of course an acceptable reason for not showing up on those days. Furthermore, all courses open for registration approximately four months before start. This can hopefully also provide new students a chance to register for courses during the year. We look forward to your registrations.


Welcome to Human-Centered Artificial Intelligence

Organizer: Niels van Berkel

Lecturers:

ECTS: 3

Date/Time: Autumn 2024

Deadline: 11 August 2024

Max no. Of participants: 20

Description: Artificial Intelligence has experienced a tremendous increase in attention in recent years across all sectors in society ranging from health, transportation, finance, construction, entertainment among others. Taking an optimistic view, Ben Schneiderman envisions,” computing devices that dramatically amplify human abilities, empowering people and ensuring human control.” He proposes that, “Human-Centered AI (HCAI), enables people to see, think, create, and act in extraordinary ways, by combining potent user experiences with embedded AI support services that users want.” Taking departure in this view, we explore prevailing research and discuss the important issues related to how AI system predict and monitor and adapt to the user. Research themes include the staples of HCI such as task performance and usability/experience and HAI-specific concerns about transparency, explainability, predictability, user control, and ethical implications.

HCI research is a fundamentally interdisciplinary field that is growing and rising to the challenges and opportunities with AI. In this course, participants will learn a concise history of the topic of Artificial Intelligence, understand the basic technical terms and techniques, and will gain an overview of broad topics of current human-centered AI research. Examples from autonomous transportation, voice interfaces, robotics, public information systems and others provide a deeper understanding of the state-of-the-art research and design of HAI. Students will learn how to critically examine HAI research articles identifying the strengths and weaknesses and possible future directions. Students will consider their current research and how themes of this course relate to their work.

 

Prerequisites:  Students should be familiar with the basic methods and practice from Human-Computer Interaction, Computer-Supported Cooperative Work, or similar fields.

Important information concerning PhD courses: We have over some time experienced problems with no-show for both project and general courses. It has now reached a point where we are forced to take action. Therefore, the Doctoral School has decided to introduce a no-show fee of DKK 3000 for each course where the student does not show up. Cancellations are accepted no later than 2 weeks before start of the course. Registered illness is of course an acceptable reason for not showing up on those days. Furthermore, all courses open for registration approximately four months before start. This can hopefully also provide new students a chance to register for courses during the year. We look forward to your registrations.

Welcome to Applications of AI to Modern Data Management Systems

Organizer: Professor Torben Bach Pedersen

Lecturers: Professor Wolfgang Lehner

ECTS: 2

Date/Time: 7-9 October 2024

Deadline: 16 September 2024

Max no. Of participants: 15

Description:

The course will cover applications of AI to modern data management systems, including learned indexes, learned optimizers, and other state of the art applications of modern AI to design of data management system or direct replacement of their components. The course will cover several types of systems, e.g., both relational database management systems and other data management/data processing systems.

Learning objectives:

The objective of the course is to provide students with a working understanding of how AI can optimize the internal workings of modern data management systems and how they can apply AI in their own data management system research.

Prerequisites:

Bachelor and master degrees in computer science or software engineering, including knowledge on machine learning and data management as introduced in typical undergraduate courses.

Important information concerning PhD courses: We have over some time experienced problems with no-show for both project and general courses. It has now reached a point where we are forced to take action. Therefore, the Doctoral School has decided to introduce a no-show fee of DKK 3000 for each course where the student does not show up. Cancellations are accepted no later than 2 weeks before start of the course. Registered illness is of course an acceptable reason for not showing up on those days. Furthermore, all courses open for registration approximately four months before start. This can hopefully also provide new students a chance to register for courses during the year. We look forward to your registrations.

Welcome to Data and Machine Learning Operations (DataOps and MLOps)

Description:  The course will cover central concepts, methods, techniques, and tools within DataOps and MLOps. Topics include Data Augmentation, Labeling, Cleaning, Pre-processing, Quantifying the Data Quality, Lifecycle, machine learning model deployment, ML pipeline orchestration, monitoring and maintenance (via updating with transfer learning OR retraining) in production, ensemble algorithms, and technical infrastructure.

Prerequisites:  Bachelor and master degrees in computer science or software engineering, including knowledge on machine learning and data management as introduced in typical undergraduate courses, as well as significant practical experience with these topics.


Organizer: Christian Thomsen

Lecturers: Alexandros Nanopoulos, University of Hildesheim

ECTS: 2

Date/Time: 4 - 6 September 2024

Place: Selma Lagerlöfs Vej 300, room 0.2.90

Zipcode: 9220

City: Aalborg

Deadline: 15 August 2024

Max no. Of participants: 15

Important information concerning PhD courses: We have over some time experienced problems with no-show for both project and general courses. It has now reached a point where we are forced to take action. Therefore, the Doctoral School has decided to introduce a no-show fee of DKK 3000 for each course where the student does not show up. Cancellations are accepted no later than 2 weeks before start of the course. Registered illness is of course an acceptable reason for not showing up on those days. Furthermore, all courses open for registration approximately four months before start. This can hopefully also provide new students a chance to register for courses during the year. We look forward to your registration.

Welcome to Knowledge Graphs and Semantic Web technologies

Organizer: Daniele Dell’Aglio, Matteo Lissandrini

Lecturers: Ernesto Jiménez-Ruiz, University of London, UK

ECTS: 2

Date/Time: 25-26 April 2024

Deadline: 4 April 2024

Max no. Of participants: 20

Description: The course will provide a practical combination of data management technology, knowledge representation and artificial intelligence. More specifically, the course will cover the following points:

(i) Knowledge Graphs and Knowledge representation. Although they are getting increased popularity, knowledge graphs are not new. Knowledge graphs have their roots in the areas of knowledge representation and Semantic Web. This course will provide an overview of ontology-based knowledge graphs following World Wide Web Consortium (W3C) standards.

(ii) Inference and Reasoning Engines. Reasoners are an important component of the Semantic Web, as they enable the inference of (implicit) knowledge from known axioms, rules and facts. Graph databases, in particular triplestores, are key engines to perform efficient reasoning over knowledge graphs.

(iii) Ontology and entity alignment. Ontology alignment is key to enable interoperability in the Semantic Web. Ontology alignment is also a useful technique in classical data integration, especially when dealing with the semantic heterogeneity problem.

(iv) Applications to Data Science. Gaining semantic understanding of arbitrary data sources (e.g., tabular data in the form of csv files) will be very valuable for data integration, data cleaning, data mining, machine learning and knowledge discovery tasks. For example, understanding what the data is can help assess what sorts of transformations are appropriate on the data.

(v) Machine learning and knowledge graphs. Machine learning and knowledge graphs (and ontologies) can complement each other. Machine learning approaches, especially neural networks, are typically noise tolerant and can generalize the inference power of ontologies and knowledge graphs. On the other hand, ontologies and knowledge graphs provide sound reasoning capabilities and can provide enhanced explainability to (black-box) machine learning models.

Prerequisites: Familiarity with python or Java.

Important information concerning PhD courses: We have over some time experienced problems with no-show for both project and general courses. It has now reached a point where we are forced to take action. Therefore, the Doctoral School has decided to introduce a no-show fee of DKK 3000 for each course where the student does not show up. Cancellations are accepted no later than 2 weeks before start of the course. Registered illness is of course an acceptable reason for not showing up on those days. Furthermore, all courses open for registration approximately four months before start. This can hopefully also provide new students a chance to register for courses during the year. We look forward to your registrations.


Welcome to Big Data Integration

Organizer: Matteo Lissandrini, Katja Hose

Lecturers: Giovanni Simonini, University of Modena and Reggio Emilia (Italy)

ECTS: 2

Date/Time: 6-7 May 2024

Deadline: 15 April 2024

Max no. Of participants: 20

Description: The course aims at illustrating recent advancements in the field of big data integration from both the practical and methodological perspective. In particular, the focus will be on tools and techniques for large and heterogenous datasets, such as data lakes and open data. The main tackled topics will be: (i) Data discovery; (ii) Entity Resolution, i.e., the task of identifying and integrating records that refer to the same real-world entity in different datasets when an explicit identifier is not provided; (iii) data preparation, i.e., the set of preprocessing operations performed to transform the data at the structural and syntactical level.

Prerequisites:  Familiarity with a programming language.

Learning objectives: Students will learn core techniques and technologies for the tasks of (i) Data discovery; (ii) Entity Resolution; (iii) data preparation.

Important information concerning PhD courses: We have over some time experienced problems with no-show for both project and general courses. It has now reached a point where we are forced to take action. Therefore, the Doctoral School has decided to introduce a no-show fee of DKK 3000 for each course where the student does not show up. Cancellations are accepted no later than 2 weeks before start of the course. Registered illness is of course an acceptable reason for not showing up on those days. Furthermore, all courses open for registration approximately four months before start. This can hopefully also provide new students a chance to register for courses during the year. We look forward to your registrations.

Welcome to Principles of Data Visualisation and storytelling

Organizer: Gabriela Montoya

Lecturers: Luis-Daniel Ibáñez (Lecturer at University of Southampton, UK)

ECTS: 2

Date: 14-15 May 2024, Time: 8:45 -16:45 both days.

Deadline: 23 April 2024

Venue: Selma Lagerløftvej 300, room 02.90

Max no. Of participants: 25

Description: For most tasks where we collect and/or analyse data, the ability to visualise what we are doing is critical for making sense to yourself and your collaborators. Appropriate visual and narrative support of the results of the analysis is even more important for communicating (and convincing!) other stakeholders such as grant decision makers and potential investors.       

In this course we will go beyond how to use a library to generate a chart and learn how to choose the appropriate chart depending on what we want to highlight and how to structure our visuals to create a compelling data story.  To do so, we will delve into the principles of human perception and study the narrative patterns that we can apply to tell a story with data. 

The assessment of the course includes an individual design and implementation of a short data story (3-5 screens) and short report (2 pages) justifying the narrative pattern and chart choices.

Knowledge will be put into practice by designing and developing a short data story in a theme of your choice.

Prerequisites: Previous experience with a data analysis software (R, MATLAB, any Python-based, Excel) 

Important information concerning PhD courses: We have over some time experienced problems with no-show for both project and general courses. It has now reached a point where we are forced to take action. Therefore, the Doctoral School has decided to introduce a no-show fee of DKK 3000 for each course where the student does not show up. Cancellations are accepted no later than 2 weeks before start of the course. Registered illness is of course an acceptable reason for not showing up on those days. Furthermore, all courses open for registration approximately four months before start. This can hopefully also provide new students a chance to register for courses during the year. We look forward to your registrations.