Description:
This course gives an introduction to probabilistic graphical models, which play an important role in modern machine learning and artificial intelligence. Various types of graphical models are introduced, and their use in typical application domains illustrated (e.g., Hidden Markov Models for biological sequence analysis, Markov Random Fields for Image Processing). The course emphasizes both modeling and learning aspects of graphical models. With regard to modeling, we discuss underlying probabilistic (independence) assumptions that need to be made when choosing a particular type of model for a given application domain. Also limitations imposed by the computational complexity of inference tasks for a given model are considered. With regard to learning, the most important paradigms for learning graphical models from data are explained, and exemplified by typical learning tasks.

Outline of contents:

Basics of probability
- Joint probability distributions
- Conditional probability and independence

Directed and undirected graphical models
- Bayesian and Markov networks, factor graphs
- Graph structure and independence relations: d-separation and the Hammersley-Clifford theorem

Latent variable models
- Hidden Markov models
- Unsupervised learning
- T
emplate models
- Plate representation

Learning and inference
- Structure learning
- Parameter estimation
- Supervised learning
- Bayesian learning

Organizer: Thomas D. Nielsen

Lecturers: Thomas D. Nielsen, Manfred

ECTS: 2.0

Time: 17 and 19 May 2022

Place:  SLV 300 i lokale 0.2.13 fra kl. 9.00-16.00. In-person attendance. 

Number of seats: 25

Deadline: 10 April 2022

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 3.000 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.




Description:
The course provides an introduction to incremental machine learning algorithms (IMLA) with an emphasis on models and applications for time-series problems. The initial lectures motivate the need for IMLA along with with an overview of the various categories of IMLA like continual learning, online learning etc. Following that, incremental variants of the standard machine learning algorithms like linear regression, SVM and random forest algorithms will be studied in detail. Other topics that would be covered in the course include IMLA for neural networks, pre-processing and post-processing tools for incremental models, concept drift, and catastrophic forgetting. The course will have an emphasis on time-series applications.

Prerequisites:
Bachelor and master’s degree in computer science or software engineering, including knowledge on machine learning and data management as introduced in typical undergraduate courses

Organizer: Professor Torben Bach Pedersen and Assistant Professor Nguyen Ho. 

Lecturers: Research Scientist Seshu Tirupathi, IBM Ireland

ECTS: 2.0

Time: May 9-12, 2022: 8:30-12:00 

Place: AAU Aalborg

Number of seats: 15

Deadline: April 18, 2022

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 3.000 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.

Description:
Deep learning on graphs has attracted significant interest recently, where graph neural networks (GNN) has achieved remarkable success in a large rank of graph-structured data domains, including e-commerce, traffic, drug design, friendship recommendation, and so on. However, most deep learning studies on graphs focus on (semi-) supervised learning scenarios, which require sufficient labeled data for effective network training. Their performance can be seriously degraded when labels are extremely limited. To address the shortcomings of (semi-) supervised learning, self-supervised learning (SSL) provides a promising learning paradigm that reduces the dependence on manual labels. Different from SSL on other domains like computer vision and natural language processing, SSL on graphs has an exclusive background, design ideas, and taxonomies. In this course, we will present a timely and comprehensive introduction of the existing approaches which employ SSL techniques for graph data. In doing so, we construct a unified framework that mathematically formalizes the paradigm of graph SSL, and then we will also briefly introduce our several recent works on SSL based GNN with extremely limited labels. Finally, we will discuss the remaining challenges and potential future directions in this research field.

Prerequisites:
The course requires basic knowledge of machine learning, especially, deep learning.

Organizer: Chenjuan Guo & Miao Zhang

Lecturers: Shirui Pan

ECTS: 3.0

Time: 9:00-12:00 on 23, 24, 25, 27 May 2022

Place: Zoom

Number of seats: 60

Deadline: 2 May 2022

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 3.000 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.

Description:
The course will address one of the most topical issues of our time, namely the widespread digitalisation of society, businesses, and public agencies. When digitalisation becomes even more radical change it gets referred to as digital transformation. In this course we will cover this through online lectures and interactive sessions with the lecturer and we will exercise the lectured material through group work. Based on this the participants will report on their learning in a small report (argumentative essay). The topics that will be covered are among others: - What is digital transformation? - Platformatisation and why it's a key driver in digitalisation? - What is blockchain? Why is it important in digitalisation? - Internet of things as a current technological trend, and how it becomes useful through artificial intelligence - How can digital transformation be conducted?

Organizer: Peter A. Nielsen

Lecturers: Carsten Sørensen

ECTS: 2.0

Time: October 2022

Place: TBA

Number of seats: TBA

Deadline: TBA

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 3.000 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.


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: Torben Bach Pedersen

Lecturer: Professor Alexandros Nanopoulos, University of Hildesheim

ECTS: 2.0

Time: September 5-7, 2022

Place:  Physical, at Department of Computer Science, AAU, Selma Lagerløfs Vej 300, 9220 Aalborg Ø, room 0.2.90

Number of seats: 15

Deadline: August 15, 2022

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 3.000 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.

Description:
This course covers the architecture and essential concepts of modern ML systems for both local and large-scale machine learning (ML). These architectures include systems for data-parallel execution (e.g., Spark, Mahout, SystemML), Parameter Servers (e.g., TensorFlow, MXNet, PyTorch), ML lifecycle systems, and the integration of ML into database systems. The covered topics focus primarily on a microscopic view of internal compilation, execution, and data management techniques, but also include a macroscopic view of entire ML pipelines. In detail, the course is composed of an introductory keynote, lectures, as well as a series of exercises. All basic concepts are augmented with pointers to recent research directions.

The course is conceptually divided into two parts:

A: Overview and ML System Internals covering topics, such as architectures, languages, operators, and various optimization techniques (runtime adaptation, parallel execution, HW accelerators, caching, data organization, etc.). 

B: ML Lifecycle Systems covering topics, such as data acquisition, cleaning, preparation, model selection, model debugging, fairness, explainability, etc.

Prerequisites:
- A general background in computer science
- Basic courses in data management or databases at undergraduate level
- Basic courses on applied ML or data mining at undergraduate level

Organizer: Katja Hose

Lecturers: Matthias Boehm (TU Graz)

ECTS: 2.0

Time: 29-30 August 2022

Place: Room 02.13 both daysTBA

Number of seats: 30

Deadline: 8 August 2022


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 3.000 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.


Description:
This course will be a hands-on course about interactive theorem proving. We will use a proof assistant to construct formal models of algorithms, protocols, and programming languages and to reason about and verify their properties. Proof assistants, like Coq or Isabelle, are tools that mechanically check human-written proofs, while assisting their users in constructing these proofs. This course will be centered around the Isabelle proof assistant. The course interleaves the introduction of various modeling and proof techniques with exercises in which students apply these techniques.

Prerequisites:
Basic mathematics, functional programming and formal semantics at the level taught for BSc/MSc students in computer Science.

Organizer: Professor (MSO) Bent Thomsen

Lecturers: Dmitriy Traytel (KU), Anders Schlichtkrull (AAU), René Rydhof Hansen (AAU)

ECTS: 2.0

Time: 16-17 May 2022, at 08.00 - 16.00

Place: room 02.90, SLV 300

Number of seats: 30

Deadline: 18 April 2022

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 3.000 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.

Description:
The Advanced Topics in Machine Learning course covers one or more selected topics of recent advances in machine learning. In particular, this year the course cover continual learning. 

Continual Learning (CL), also referred to as Lifelong or Incremental Learning, studies the problem of learning from a stream of data from changing domains, each connected to a different learning task. The objective of CL is to quickly adapt to new situations or tasks by exploiting previously acquired knowledge, while protecting previous learning from being erased. 

Teaching methods will mainly be lectures. 

Evaluation form: Hand in a short article (one page) to discuss the course content may potentially help solve the problems in the students' PhD projects.

Prerequisites:
Machine learning basics.

Organizer: Bin Yang/Miao Zhang

Lecturers: Bing Liu & Zixuan Ke, University of Illinois at Chicago (UIC)

ECTS: 2.0

Time: 15.00 to 19.00, 14 June 2022.  15.00 to 19.00, 16 June 2022. 

Place: Online 

Number of seats: 50

Deadline: 24 May

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 3.000 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.