Trustworthy Machine Learning

Machine learning has the potential for tremendous health innovations, but applying it in healthcare poses novel and interesting challenges. Data privacy is paramount, applications require high confidence in model quality, and practitioners demand explainable and comprehensible models. Ultimately, practitioners and patients alike must be able to trust these methods. In our research group on Trustworthy Machine Learning we tackle these challenges, investigating novel approaches to privacy-preserving federated learning, the theoretical foundations of deep learning, and collaborative training of explainable models.

Open Positions

  • We offer Master and Bachelor theses

If you are interested, please send an email to Michael Kamp.

Group Leader

Dr. Michael Kamp

Members

Amr Abourayya, M.Sc.

External Homepage

    2024

  • Layer-wise linear mode connectivity
    Adilova, L., Andriushchenko, M., Kamp, M., Fischer, A., & Jaggi, M.
    In The Twelfth International Conference on Learning Representations
  • Visual Computing for Autonomous Driving
    Chen, S., Gou, L., Kamp, M., & Sun, D.
    IEEE Computer Graphics and Applications, 44(3), 11–13
  • Landscaping Linear Mode Connectivity
    Singh, S. P., Adilova, L., Kamp, M., Fischer, A., Schölkopf, B., & Hofmann, T.
    In ICML Workshop on High-dimensional Learning Dynamics: The Emergence of Structure and Reasoning
  • Orthogonal Gradient Boosting for Simpler Additive Rule Ensembles
    Yang, F., Le Bodic, P., Kamp, M., & Boley, M.
    In International Conference on Artificial Intelligence and Statistics (pp. 1117–1125) PMLR

    • 2023

  • FAM: Relative Flatness Aware Minimization
    Adilova, L., Abourayya, A., Li, J., Dada, A., Petzka, H., Egger, J., et al.
    TAGML2023
  • Re-interpreting Rules Interpretability
    Adilova, L., Kamp, M., Andrienko, G., & Andrienko, N.
    International Journal of Data Science and Analytics
  • Federated Learning from Small Datasets
    Kamp, M., Fischer, J., & Vreeken, J.
    In International Conference on Learning Representations (ICLR)
    federatedlearningsmalldatasets
  • Open-source skull reconstruction with MONAI
    Li, J., Ferreira, A., Puladi, B., Alves, V., Kamp, M., Kim, M., et al.
    SoftwareX, 23, 101432
  • MedShapeNet – A Large-Scale Dataset of 3D Medical Shapes for Computer Vision
    Li, J., Pepe, A., Gsaxner, C., Luijten, G., Jin, Y., Ambigapathy, N., et al.
  • Nothing but Regrets - Privacy-Preserving Federated Causal Discovery
    Mian, O., Kaltenpoth, D., Kamp, M., & Vreeken, J.
    In International Conference on Artificial Intelligence and Statistics (AISTATS)
  • Information-Theoretic Causal Discovery and Intervention Detection over Multiple Environments
    Mian, O., Kamp, M., & Vreeken, J.
    In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI)
    mian2023information

    • 2022

  • AIMHI: Protecting Sensitive Data through Federated Co-Training
    Abourayya, A., Kamp, M., Ayday, E., Kleesiek, J., Rao, K., Webb, G. I., & Rao, B.
    In FL-NeurIPS 2022)
  • Open-Source Skull Reconstruction with MONAI
    Li, J., Ferreira, A., Puladi, B., Alves, V., Kamp, M., Kim, M. -S., et al.
    arXiv preprint arXiv:2211.14051
  • Regret-based Federated Causal Discovery
    Mian, O., Kaltenpoth, D., & Kamp, M.
    In The KDD′22 Workshop on Causal Discovery (pp. 61–69) PMLR
  • When, Where and How does it fail? A Spatial-temporal Visual Analytics Approach for Interpretable Object Detection in Autonomous Driving
    Wang, J., Li, Y., Zhou, Z., Wang, C., Hou, Y., Zhang, L., et al.
    IEEE Transactions on Visualization and Computer Graphics

    • 2021

  • Artificial Neural Networks Implementation to Predict the Solution of Nonlinear ODE System for the Application to Turbulent Combustion Modeling
    Abourayya, A.
    Master’s thesis, TU Wien
    link
  • FedBN: Federated Learning on Non-IID Features via Local Batch Normalization
    Li, X., Jiang, M., Zhang, X., Kamp, M., & Dou, Q.
    In International Conference on Learning Representations
  • Relative flatness and generalization
    Petzka, H., Kamp, M., Adilova, L., Sminchisescu, C., & Boley, M.
    Advances in neural information processing systems, 34, 18420–18432

    • 2018

  • Unveiling CO adsorption on Cu surfaces: new insights from molecular orbital principles
    Gameel, K. M., Sharafeldin, I. M., Abourayya, A. U., Biby, A. H., & Allam, N. K.
    Physical Chemistry Chemical Physics, 20(40), 25892–25900

    • 2017

  • Theoretical and DFT Analysis of the CO Adsorption Mechanism Late Transition Metal Surfaces
    Abourayya, A., Gameel, K. M., Sharafeldin, I. M., Biby, A. H., & Allam, N. K.
    NanoWorld Conference, Bosten, USA

About the INI

The Institut für Neuroinformatik (INI) is a research unit of the Faculty of Computer Science at the Ruhr-Universität Bochum. Its scientific goal is to understand the fundamental principles through which organisms generate behavior and cognition while linked to their environments through sensory and effector systems. Inspired by our insights into such natural cognitive systems, we seek new solutions to problems of information processing in artificial cognitive systems. We draw from a variety of disciplines that include experimental psychology and neurophysiology as well as machine learning, neural artificial intelligence, computer vision, and robotics.

Contact

Universitätsstr. 150, Building NB, Room 3/32
D-44801 Bochum, Germany

Tel: (+49) 234 32-28967
Fax: (+49) 234 32-14210

sekretariat@ini.rub.de
www.ini.rub.de