ESR photo Research project "Algorithmic governance for active assisted living" About the project Algorithmic decision making became enmeshed into daily life. In active assisted living data is analysed and interpreted with the intention to support people in various ways: recognizing behaviour, events, emotions, needs; creating ambient intelligence; predicting activities and proposing treatment strategies. Machine learning as prerequisite of intelligence is applied. Taking into account recent success, it can be claimed, that not only a set of specific algorithms but also a lot of example data is needed to run the learning methods. And usually those building the algorithms are not trained in law or the social sciences, while experts in discrimination law do not know how to audit modern machine learning algorithms. Further complicating matters is that even experts in computer science and mathematics often struggle with interpreting the output of many modern machine learning algorithms. Unsurprisingly assessing and guaranteeing fairness and transparency in machine learning is a wide open research and that is the topic of the PhD proposal. Progress of the project The first part of the project has been dedicated to education in the subject of machine learning, AI fairness, bias, and transparency, as well as interviews with developers at the TU Wien and a review of the use of explainable AI in audio and video-based AAL Application. The completion of the review of the state of the art, led to the decision of examining several AAL systems (fall detection system and automatic speech recognition systems) by using a pre-processing, in-processing, post-processing framework. Pre-processing experiments with fall detection include creating synthetic data to improve the human diversity in fall detection data and testing state-of-the-art solution on a more diverse dataset. In parallel, experiments on fair privacy have been conducted with ESR13. Collaborations have also been undertaken with ESR3, participating as an AI expert in his studies. Scientific publications Fairly Private: Investigating The Fairness of Visual Privacy Preservation Algorithms Sophie Noiret, Siddharth Ravi, Martin Kampel, Francisco Florez-Revuelta Fairly Private: Investigating The Fairness of Visual Privacy Preservation Algorithms Fourth AAAI Workshop on Privacy-Preserving Artificial Intelligence (PPAI-23), Washington, DC, USA, 2023 On The Nature of Misidentification With Privacy Preserving Algorithms Sophie Noiret, Siddharth Ravi, Martin Kampel, Francisco Florez-Revuelta On The Nature of Misidentification With Privacy Preserving Algorithms In Proceedings of the 15th International Conference on PErvasive Technologies Related to Assistive Environments, pp. 422-424, 2022. Bias and Fairness in Computer Vision Applications of the Criminal Justice System Sophie Noiret, Jennifer Lumetzberger, Martin Kampel Bias and Fairness in Computer Vision Applications of the Criminal Justice System In 2021 IEEE Symposium Series on Computational Intelligence (SSCI), pp. 1-8, IEEE, 2021. About the ESR Sophie received her Master's Degree in Engineering from the Ecole Centrale de Nantes (France) in 2018, with a specialty in Robotics and Embedded Systems. She has since worked in the aeronautics industry, developing software for the Rafale plane. Contact information Sophie Noiret Vienna University of Technology Computer Vision Lab Favoritenstr. 9/193-1 A-1040 Vienna, Austria Email address: snoiret@cvl.tuwien.ac.at
Fairly Private: Investigating The Fairness of Visual Privacy Preservation Algorithms Sophie Noiret, Siddharth Ravi, Martin Kampel, Francisco Florez-Revuelta Fairly Private: Investigating The Fairness of Visual Privacy Preservation Algorithms Fourth AAAI Workshop on Privacy-Preserving Artificial Intelligence (PPAI-23), Washington, DC, USA, 2023
On The Nature of Misidentification With Privacy Preserving Algorithms Sophie Noiret, Siddharth Ravi, Martin Kampel, Francisco Florez-Revuelta On The Nature of Misidentification With Privacy Preserving Algorithms In Proceedings of the 15th International Conference on PErvasive Technologies Related to Assistive Environments, pp. 422-424, 2022.
Bias and Fairness in Computer Vision Applications of the Criminal Justice System Sophie Noiret, Jennifer Lumetzberger, Martin Kampel Bias and Fairness in Computer Vision Applications of the Criminal Justice System In 2021 IEEE Symposium Series on Computational Intelligence (SSCI), pp. 1-8, IEEE, 2021.