Goal: In this project we want to contribute to the development and the deployment of human-centred Artificial Intelligent systems. The integration of our projects will ideally ensure that Artificial Intelligence will be applied in a more conscious way and more aligned with our values.
Context: Despite tremendous progress in the past few years, the field is currently facing some of the most important challenges at the interface between theoretical computer science, applied data science, and sociotechnical studies. Indeed, the most successful machine learning models (deep neural networks) are non-transparant, highly non-intuitive, and therefore difficult to understand.
Methods: Here, we join forces to tackle this problem on three different levels. First, we want to use different techniques to overcome the traditional interpretability vs learning performance trade-off. Then, we want to develop tools to interrogate true black box systems, i.e., systems where only the outcome is observed (and nothing of the internal process). Finally, we want to find efficient ways to valorise the academic insights in society through the development of recommendations and tools that can be used to train and validate trustworthy neural networks in a real-life environment.
Affiliated Research Groups:
FTI Afterwork: Back to the Future
In March, the Flanders Technology & Innovation (FTI) Festival will take place in Brussels. In the evening of the 20th of March, an FTI afterwork with Pecha Kucha-style presentations will take place. Together with other technologie experts, our researcher Carmen Mazijn takes on the challenge of pitching her message with 20 slides for 20 seconds. Register here: https://events.vub.be/fti-festival-brussel-back-to-the-future
PILAR ASAP: Surveillance Art, Dying Phones & Fake Likes
PILAR ASAP festival is an extraordinary festival that explores the boundaries of art, technology and human experience. This year the theme is The Posthuman Edition. On the 12th of March, Dries Depoortere gives a talk on Surveillance Art, Dying Phones and Fake Likes. Afterwards our research Carmen Mazijn will dive deeper into these topics with the audience. Get your tickets through this LINK.
De ethische aspecten van AI
On 8/11/23, our researcher Carmen Mazijn gave a talk about her work in Bruges. Her talk was part of the celebrations of the 40th anniversary of the school for adult education SNT.
Short video on the paper 'Exposing Bias through Inverse Design'
Explanation video on the paper presented at the ICML WS Resonisble Decision-making in Dynamic Invironments 23/7/22.
Demos&Drinks: Women in AI and Robotics Research
On 20 October 2022, our member Carmen Mazijn shared her research at the Demos&Drinks: Women in AI and Robotics Research at BeCentral.
Check out the event here.
Broodje Kennis: Kan kunstmatige intelligentie een eerlijke beslissing nemen?
Lunch lecture on 22 April 2023 (in Dutch) at deBuren in Brussels.
Check out the event and get tickets here.
Preventing biases and discrimination in / with Artificial Intelligence
Panel discussion on 1 April 2022 organized by FARI - AI for the Common Good Institute for the weKONEKT week.
Check the recording here.
This brAInfood contains an explanation of..
- 3 different fairness notions,
- what is a threshold,
- and the difference between group fairness and individual fairness
...using visualisations and a fictional scenario of an automated job application process. Read all about it here!
On 5-6 July 2022, the Brussels FARI AI Institute for the Common Good organised a conference on AI, Data and Robotics in Cities. During this event, public officials, researchers, international city representatives and citizens were invited. An Observer Committee, including our PhD candidate Carmen Mazijn, attended all sessions to identify what the priorities could be for the Brussels Capital Region (BCR). They summarized the results and made policy recommendations for the region. Read the document here.
Workshop: Does the computer give you the job you deserve?
In cooperation with Knowledge Centre Data & Society. You can find more information and where we will be holding this workshop next here.
We create a Github-repository where you can find the code to implement LUCID and LUCID-GAN. Test it out and let us know what you think: Gitub