Forschungspraxis/Master’s thesis - Communicating Uncertainty in Predictive Decision Making


Topic

The perception of Machine Learning (ML) based decision support systems as “black boxes” is a significant barrier to adoption. This is especially the case in high-stakes domains like medicine, where such systems need to earn a high degree of trust before their usage is approved. Recent publications in the field of explainable AI (XAI) suggest that relaying a system’s uncertainty to the user is an important aspect of transparency and has the potential to increase trust. However, the perceived trustworthiness of a system is likely to be dependent on a range of factors, including the visual presentation of the model’s uncertainty and the user’s personal background. The goal of this Forschungspraxis or Master’s thesis is to examine these factors by conducting a user study that draws on theory from the fields of uncertainty visualization and XAI.

The student work will be carried out in conjunction with the CellFace project, which investigates the potential of computer vision and machine learning techniques for the task of automated blood cell diagnosis. You can read more about CellFace here.

Requirements

  • Prior experience with (or a strong interest in) human-computer interaction methodologies and data visualization
  • Experiences in Python and Git
  • Capacity for teamwork

For a more in-depth introduction to the relevant research topics, see:

Sacha, Dominik, Hansi Senaratne, Bum Chul Kwon, Geoffrey Ellis, and Daniel A. Keim. 2016. “The Role of Uncertainty, Awareness, and Trust in Visual Analytics.” IEEE Transactions on Visualization and Computer Graphics 22 (1): 240–49. https://doi.org/10.1109/TVCG.2015.2467591.

Arshad, Syed Z., Jianlong Zhou, Constant Bridon, Fang Chen, and Yang Wang. "Investigating user confidence for uncertainty presentation in predictive decision making." In Proceedings of the Annual Meeting of the Australian Special Interest Group for Computer Human Interaction, pp. 352-360. 2015.

Supervisors

Alice Hein, M.Sc. and Stefan Röhrl, M.Sc.

Chair for Data Processing

Contact Information

Alice Hein, M.Sc.

Chair for Data Processing

TUM Department of Electrical and Computer Engineering

Technical University of Munich

 

Arcisstr. 21, 80333 Munich

Room Z942

alice.hein@tum.de