Julia Grabinski
Researcher & PhD
Candidate
Chair for Machine Learning,
University of Mannheim Data and Web Science Group,
School of Business Informatics and Mathematics
B6, 26, Room C0.02 68159 Mannheim
Email: julia.grabinskiuni-mannheim.de
I'm a Ph.D. candidate at the Chair for Machine Learning supervised by Prof. Dr.-Ing Margret Keuper.
Research Interests
- Signal Processing in modern Computer Vision
- Robustness and Reliability of Deep Learning Models
Publications
J Grabinski, P Gavrikov, J Keuper, M Keuper: Robust Models are less Over-Confident, https://proceedings.neurips.cc/paper_files/paper/2022/file/ff887781480973bd3cb6026feb378d1e-Paper-Conference.pdf, NeurIPS 2022
J Grabinski, J Keuper, M Keuper: Aliasing and adversarial robust generalization of CNNs, https://link.springer.com/article/10.1007/s10994-022-06222-8, ECML 2022
J Grabinski, S Jung, J Keuper, M Keuper: FrequencyLowCut Pooling – Plug & Play against Catastrophic Overfitting, https://arxiv.org/pdf/2204.00491.pdf, ECCV 2022
J Grabinski, J Keuper, M Keuper: Aliasing coincides with CNNs vulnerability towards adversarial attacks, https://openreview.net/pdf?id=vKc1mLxBebP, AAAI-22 Workshop on Adversarial Machine Learning and Beyond