BIASlab
Research focus
We are interested in modern AI methods that support automated design of signal processing systems, in particular in the context of audio processing algorithms. For instance, if I have a problem with (the signal processing of) my hearing aid when I am at a cocktail party, I want to fix it right there so I can stay and enjoy the party. Today, there are no proper tools that support the end user in fine-tuning his hearing aid. We aim to develop such a support tool, and our methods are inspired by theories about learning and adaptation in the fields of (Bayesian) machine learning and computational neuroscience. In particular, we derive inspiration from recent work in neuroscience that describes how brains perceive, learn and design their algorithms for speech and object recognition, navigation, etc. In our team, we are building a very efficient Bayesian machine learning toolbox ourselves and ‘eat our own dogfood’ by putting ourselves in complex acoustic situations and use our own tools to tune our audio processing algorithms. This leads to new demands on the toolbox, which drives our next research steps.
In short, we work on (Bayesian) machine learning methods with applications to automated design of signal processing algorithms. Our methods are just as easily applicable though to design problems in related fields such as control, biomedical and communications engineering.
Courses
The courses offered by the BIASlab tie in perfectly with our research focus. In BIASlab we focus on Bayesian machine learning which you will learn in the specialization course of the SPS master. As a background we would highly recommend you to follow the fundamentals of machine learning course, in which we teach you the basics of the field of machine learning. As all machine learning requires proper software engineering, we also learn you how to properly write software for machine learning applications.
| Course code | Course name | Level | Time Slot |
|---|---|---|---|
| 5XSL0 | Fundamentals of Machine Learning | Bachelor/Advance Elective | Q4 |
| 5SSD0 | Bayesian Machine Learning and Information Processing | Master Specialization | Q2 |
| 5ARA0 | Software Engineering for Artificial Intelligence | Master Elective | Q3 |
Graduation
To finish the Master studies you have to enroll in a Graduation Project. You can discover some of our projects in the Master marketplace. There you can find the latest information regarding the registration as well as an extended description of the active projects within BIASlab. Just remember that once you are logged in the Marketplace portal, you can filter the list of projects by choosing the capacity group. Choose SPS in the Research group option, and select Bert de Vries as the Responsible staff. If you have any questions regarding one of our projects, please do not hesitate to contact the responsible for the project you are interested in.
Internships
BIASlab cooperates with industrial partners that frequently have internships available for graduate students with a background in machine learning. Get in touch with us to discuss the available options for an internship. We are happy to shape the internship according to your own wishes and interests.
Contact
More information about BIASlab can be found at our site. Do you have any questions regarding the information above? Please do not hesitate to contact prof. dr. ir. Bert de Vries by sending him a mail at bert.de.vries@tue.nl.
Collaboration partners
We are grateful to work with our collaborative partners:
