Welcome to Topics in Audio and Acoustic Signal Processing (2021)

Description: Audio and acoustic signal processing deals with the science and practice of signal processing for signals of an acoustic origin. As such, it is the discipline behind such marvels of mordern society as hearing aids, streaming services, smart phones, and smart speakers. The course will cover recent advances within the field,  including including robust acoustic beamforming and localization, speech enhancement, sound zones, noise statistics estimation, wireless acoustic sensor networks, active noise cancellation, and robot audition. These are based on methodological developments wherein disciplines such as machine learning and convex optimization are integrated into signal processing. The course will feature a number of lectures and hands-on exercises.

Organizer: Professor Mads Græsbøll Christensen - mgc@create.aau.dk

Lecturers: Professor Mads Græsbøll Christensen - mgc@create.aau.dk, Associate Professor Jesper Rindom Jensen - jrj@create.aau.dk

ECTS: 2.0

Time: November 22-26 2021

Place: Aalborg University

Zip code: 
9220

City: Aalborg

Number of seats: 25

Deadline: November 01 2021


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.
Welcome to Introduction to Bayesian Learning (2021)


Description: This short course is dedicated to Bayesian learning. The Bayesian approach to Machine Learning adopts a radically different viewpoint to the more “standard” approaches. The latter are based on models that are parametrized in terms of a set of unknown parameters, which are considered to be deterministic variables. That is, each parameter, although unknown, it corresponds to a specific fixed value. 

In contrast, in the Bayesian world, the unknown parameters are treated as random variables.  This was a revolutionary idea at the time it was introduced by the mathematician and philosopher Bayes and later on used by the great mathematician Laplace. Even now, after more than two centuries, it may seem strange to assume that a physical phenomenon/mechanism is controlled by a set of random parameters.

However, there is a subtle point here. Treating the underlying set of parameters as random variables, we do not really imply a random nature for them. The associated randomness, in terms of prior distribution, encapsulates our uncertainty about their values, prior to receiving any observations. Stated differently, the prior distribution represents our belief about the different possible values, although only one of them is actually true. From this perspective, probabilities are viewed in a more “open-minded” way, that is, as measures of uncertainty.

The outline of this course is as following:

 

  • Maximum Likelihood and Maximum a-Posteriori Estimators: A Revision
  • The Bayesian Approach: Basic Concepts
  • The Bayesian Approach to Linear Regression: The Gaussian Case
  • The Evidence Function and Occam's Razor Rule
  • The Laplacian Approximation and the Evidence Function
  • Latent Variables and the Expectation-Maximization (EM) Algorithm
  • Linear Regression and the EM Algorithm
  • Gaussian Mixture Models  and the k-Means Algorithm
  • The Lower Bound Interpretation of the EM Algorithm
  • Exponential Family of Distributions
  • Variational Approximation and the Mean Field Approximation Concept: A Discussion



Organizer: Jan Østergaard

Lecturers: Sergios Theodoridis

ECTS: 2.0

Time: 10-14 May 2021. From 9:00 - 12:00.

Place: Aalborg University

Zip code: 
9220

City: Aalborg

Number of seats: 50

Deadline: 19 April 2021


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.
Welcome to (CASPR) Summer School (2021)


Description: Hearing assistive devices (HADs) are ubiquitous. They include, for example, devices such as headsets for speech communication in noisy environments (air plane crews, emergency/rescue teams, combat soldiers, police forces, etc.), headsets for office use, gaming, etc., and hearing care systems, e.g. hearing aids and cochlear implants.

The course consists of lectures and hands-on exercises, which allow the course participants to understand in-depth the technical problems related to HADs and their potential solutions. The course is multi-disciplinary, with a focus on application of theoretical results to real-world problems, and practical do's and don'ts.

The course has three main parts. The first part is a short introductory part, which lays the foundation for the rest of the course, covering fundamental topics such as auditory perception (normal and impaired hearing) and a discussion of the basic principles of HADs. The second part provides an overview of fundamental signal processing problems encountered in HADs, and an in-depth treatment of state-of-the-art solutions. These include methods for beamforming and noise reduction, direction-of-arrival estimation, voice activity detection, feedback control, hearing loss compensation, etc. Furthermore, an overview is given of methodologies for evaluating HADs with a particular focus on methods for intelligibility assessment and estimation. The third part of the course presents emerging technologies for hearing assistive devices, including machine learning techniques for processing of speech in noise, audio-visual signal processing, intelligibility prediction, user-aware/symbiotic signal processing, methods for own-voice processing, etc. While the course focuses on the HAD application, many of the discussed techniques are general and find use in the much broader field of general sound processing.

The course also bridges the gap between theoretical background and practical/robust application. The course is a one-week concentrated course to be held in the Summer, 2021. The course involves course preparation (approximately 1 ECTS), course presence (1 ECTS), assignment finalization and hand-in (1 ECTS)

Prerequisites: Basic knowledge of statistical signal processing, stochastic processes, and linear algebra. Familiarity/handy with Matlab/Python.

The course schedule is still under preparation.

Course Fee: Free for PhD students and postdocs. The fee is 7500 DKK (1000 EUR) for industrial participants.

For more information about the Summer School, feel free to contact the organizers:
Prof. Jesper Jensen (jje@es.aau.dk), Prof. Zheng-Hua Tan (zt@es.aau.dk), and Prof. Jan Østergaard (jo@es.aau.dk).

The course will be given following the guidelines of the Danish Health Authorities on the COVID-19, and will take place with physical presence of (and physical distance between) the course participants. In case the Covid-19 situation prevent us from having physical presence, an alternative online version of the course will be introduced.



Organizer: Professor Jesper Jensen - jje@es.aau.dk

Lecturers: Prof. Jesper Jensen, AAU, Prof. Jan Østergaard, AAU.
Prof. Zheng-Hua Tan, AAU., Dr. Meng Guo, Oticon (invited talk), Dr. Asger Heidemann Andersen, Oticon (invited talk), Dr. Carina Graversen, Eriksholm (invited talk), Poul Hoang (AAU), Payam Shahsavari (AAU), Dr. Daniel Michelsanti (AAU), Dr. Iván López Espejo

ECTS: 3.0 

Time: 17-21 May 2021

Place: Aalborg University, Fredrik Bajers Vej 7b, 9220 Aalborg, Denmark.

Zip code: 
9220

City: Aalborg

Number of seats: 30

Deadline: 26 April 2021


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.
Welcome to Control and Optimization (2021)


Description: Optimal control is the problem of synthesizing controllers for dynamic systems such that a certain performance function is minimized. Today, the stress is on developing efficient numerical methods for solving a class of optimal control problems (herein convex optimization). Optimal control finds its application not only in engineering, but also in economics, biology and logistics.

The course has two main parts, which in headlines are: "Optimal Control" and "Optimization".

In the first part of the course we concentrate on the foundation of optimization aiming at the characterization of necessary conditions for local optimality given by the celebrated Karush-Kuhn-Tucker theorem. We then specialize to convex optimization techniques including Dual Decomposition and the Alternating Direction Method of Multipliers (ADMM), as they provide tangible methods for solving (convex) optimization problems. 

The second part of the course will be devoted to optimal control in the form of Model Predictive Control (MPC). We will focus on MPC for linear dynamical systems with quadratic performance function, and show how techniques from optimization enable us to synthesis controllers in the MPC scheme.

Prerequisites: Knowledge of linear algebra and basic calculus.



Organizer: John-Josef Leth

Lecturers: John-Josef Leth, Jan Østergaard

ECTS: 2.0

Time: 19 - 23 April 2021

Place: Aalborg University

Zip code: 
9220

City: Aalborg

Number of seats: 30

Deadline: 29 March 2021


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.
Welcome to Methods for Data Collection with Human Subjects (2021)



Description: Much of technical scientific research has as an ultimate goal to develop technological devices that will be used by people. Either as services, or products that may end up being an integral part of peoples’ lives. In order to investigate if a given technology can be used in a given context, experiments with human subjects are indispensable. These types of experiments can be time consuming and expensive and if not properly designed and executed the results may not be reliable.

This course takes both a theoretical and practical approach to data collection from human subjects, in order to avoid mistakes and common pitfalls while improving the reliability of the results. From an understanding of cognitive and perceptual processes of human interaction with the environment, we will investigate implications of these processes in different data collection scenarios.

The following scenarios will be presented based on the research activities of the section:

  1. Evaluation of sound environments and interactive control (sound zones).
  2. Hearing aid aided performance ratings (BEAR).
  3. EEG differences between active and passive listening.

The course consist of lectures and practical work, where you will design and participate in experiments.


Prerequisites: Knowledge of basic data analysis models, such as CHI-square, T-tests, ANOVA is an advantage, but not strictly required. Programming in any language capable of making simple user interfaces.


Learning Objectives:

  • Understand ethical and practical implications of using humans as test subjects.
  • Understand and learn to work with rules and regulations for personal data (GDPR).
  • Understand how to avoid common mistakes in the experimental design.
  • Understand and know the importance of familiarisation and training.
  • Understand and learn to apply methods for balancing levels of independent variables.
  • Understand and be able to work with variability in subject responses.
  • Be able to implement methods for data collection that can minimise subject bias.
  • Be able to implement and work with difference judgements, rating scales and forced choice methods.
  • Be able to derive scales based on human responses


Organizer: Associate Professor Rodrigo Ordoñez - rop@es.aau.dk

Lecturers: Associate Professor Rodrigo Ordoñez and Lars Bo Larsen

ECTS: 2.0

Time: 18-19 March and 15-16 April 2021

Place: Aalborg University

Zip code: 
9220

City: Aalborg

Number of seats: 30

Deadline: 25 february 2021


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.
Welcome to State and Parameter Estimation in Nonlinear Dynamic Stochastic Systems (2021)


Description: In applications within modeling, estimation, control and detection state space models are used. Normally the whole state can not be measured but only some function of the state is measured and perhaps also with substantial measurement noise. In at least the above applications it is necessary to estimate the whole state from the measurement. Exactly this is the topic for the course. For linear discrete time stochastic systems there exist the Kalman filter solution which to some extend is included in the master curriculum. The topic for this course is the advanced methods for nonlinear and continuous time systems and also to include parameter estimation. The purpose of this PhD course is to give the participant a comprehensive knowledge on both basic and more advanced aspects of state and parameter estimation. The goal is to enable the students with knowledge and tools for stochastic modeling of physical systems. The participant should be able to apply software for state and parameter estimation for the model structures. The software used are programs from MATLAB. Topics include among others Kalman filters (KF), Extended KF, Unscented KF, continuous discrete filtering, stochastic differential equation (SDE), parameter estimation by extending the state or maximum likelihood (ML), particle filter (PF). The course is evaluated as passed/not passed. In order to pass the student must be actively participating and deliver written solutions to the exercises which must be accepted by the lecture. 

Organizer: Torben Knudsen

Lecturers: Torben Knudsen

ECTS: 3.0

Time: 3-7 May 2021

Place: Aalborg University

Zip code: 
9220

City: Aalborg 

Number of seats: 25 

Deadline: 01 April 2021


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.

Welcome to Deep Learning - Electrical and Electronic Engineering (2021)


Please sign up for this course on the link below:

https://phd.moodle.aau.dk/course/view.php?id=1553
Welcome to Stochastic Safety Control (2021)


Description: The course comprises three parts: 

BackgroundIn this part, we will discuss Markovian processes, both continuous and discrete (Markov chain). We will discuss the definition of conditional expectations and martingale theory to optional sampling theorems. 

Safety and Control: In this part, we will first discuss safety. Stochastic safety deals with the computation of the probability that a system hits a forbidden state. This notion is related to fault detection, the detection of cyber-attacks, obstacle avoidance in robotics. To this end, we will use the idea of the occupation measure. That is the probability that a realization of a process belongs to a given set. We will show how to design a safe controller that reaches a target set with a “very low probability” of hitting a forbidden set. 

Numerical Methods: The safety and control are translated into a problem of certificates of positivity. The task here is to find a polynomial that is positive in a given set. There are powerful tools available leaning upon sums of polynomials. The sum of squares can be solved by means of semi-positive programming, e.g., in Yalmip (Matlab).


Organizer: Rafael Wisniewski

Lecturers: Rafael Wisniewski (ES, AAU), Manuela L. Bujorianu, Maritime Safety Research Center, Department of Naval Architecture, University of Strathclyde, Scotland, UK

ECTS: 3.0

Time: 12-16 April 2021

Place: Aalborg University

Zip code: 
9220

City: Aalborg

Number of seats: 20

Deadline: 22 March 2021


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.