Maike Rees

Hi! My name is Maike, I am 27 years young and just started my PhD to research for a better tomorrow. Let me introduce myself by highlighting the main stages of my career so far.

My Bachelor was about Media, Communication and Computer Science where I gained a very good and hands-on training in classic software engineering. I included a semester abroad in Canada (close to Toronto) and wrote my Bachelor’s thesis in beautiful San Francisco about Comparison of Machine Learning Techniques for Anomaly Detection in Autonomous Driving at the Volkswagen Group of America Electronics Research Lab.

I did my Master in Computer Science at the KIT in Karlsruhe with a focus on artificial intelligence (and a bit of software engineering). During my master studies, I gained experience in industrial, academic and interdisciplinary research: at the R&D department of Porsche in Weissach I held a working student position about robust autonomous driving and I am especially proud of our Classification by Components Paper that was accepted at NeurIPS 2019 (find more details below); at the Institute of Applied Materials (IAM-WK) at the faculty of mechanical engineering at the KIT I worked as a student assistant, researching AI for quality assurance. Even more interdisciplinary experience origins from my Master’s thesis that I wrote about the Analysis of the Explainability of a Deep Learning Algorithm for the Localization of Ventricular Ectopic Foci based on Body Surface Potential Maps at the Institute of Biomedical Eengineering (IBT) at the faculty of electrical engineering and information technology at the KIT.

In 2021, I started my PhD at the DKFZ (German cancer research center) in the group of Lena Maier-Hein, called Computer Assisted Medical Interventions (CAMI) where I work on computer vision and machine learning with uncertainty prediction to make surgeries safer.

I am interested in computer vision (especially concerning non-RGB images), explainable AI (XAI), safe AI, robust AI, uncertainty in AI, scientific writing, clean code (especially in research opposed to classic software engineering) and teaching. Don’t hesitate to shoot me a message and discuss these topics with me :-)

In my free time I swim, run, dance (salsa), lift weights, read, surf and do yoga. I also volunteer extensively at my Lifesaving Club (DLRG - Deutsche-Lebens-Rettungs-Gesellschaft), especially teaching kids how to swim, teaching students the skills of life saving and teaching trainers how to teach their students. I also enjoy traveling a lot.

Publications

Classification-by-Components: Probabilistic Modeling of Reasoning over a Set of Components,
Sascha Saralajew, Lars Holdijk, Maike Rees, Ebubekir Asan, Thomas Villmann; NeurIPS 2019
>Paper >Poster >Code

Robustness of Generalized Learning Vector Quantization Models Against Adversarial Attacks,
Sascha Saralajew, Lars Holdijk, Maike Rees, Thomas Villmann; WSOM 2019
>Paper >Code

Prototype-based Neural Network Layers: Incorporating Vector Quantization,
Sascha Saralajew, Lars Holdijk, Maike Rees, Thomas Villmann,
>Technical Report >NeurIPS 2018 Workshop Poster

Modular Novelty Detection System for Driving Scenarios,
Maike Rees, Melanie Senn, Pratik P. Brahma, Astrid Laubenheimer; CERC 2019
>Paper

Talks

Machine Learning Student Talk of the University of Applied Sciences Karlsruhe in October 2018 about Modular Novelty Detection System for Driving Scenarios

Visit me on github, Google Scholar or LinkedIn.