Job description:
Deep learning (DL) has achieved remarkable results for perceptual tasks within the last decade. However, DL-based perception often lacks sufficient robustness for real-world applications, as exemplified by the existence of adversarial examples and the fragility in face of natural distortions not foreseen during training. Besides, there is growing evidence that DL-based perception works differently than human perception on a fundamental level, e.g. relying overly strong on texture cues and on brittle characteristics of the training data.
In this PhD, we want to work on fundamentally new methods for DL, for instance new network architectures, new training procedures, or new regularization schemes.
The results should be published at the top-tier machine learning venues.
Qualifications:
* Personality: Communicative and team player
* Working Practice: Independent, motivated to work in an interdisciplinary and international team
* Experience and Knowledge: With deep learning frameworks (TensorFlow, PyTorch, etc.), basic knowledge of machine learning and deep learning, strong programming skills, in particular Python and strong mathematical background
* Languages: Very good in English (written and spoken)
* Education: Excellent degree (Master) in computer science, mathematics or related fields with excellent marks.
Additional Information:
Please refer to http://smrtr.io/3jL_w for further information and how to apply.
Looking forward to your application!
Jan Hendrik Metzen