Master Thesis / Internship - Deep learning with layer-local objectives

Your Job:

In this project, we'll probe representations in hidden layers of pretained networks, specifically ViT- and ConvNext-type architectures, to reverse engineer the types of objectives that might optimally lead to these representation. We then aim to understand whether the obtained objectives, applied in a layer-local manner, lead to performant deep network learning.

Your tasks in detail:

• Implement a flexible data loading framework capable of incorporating multiple datasets
• Develop strategy to decode image-to-image as well as image-to-label tasks from hidden representations
• Train decoders for multiple datasets / datatypes on the representations in hidden layers of pretrained networks (ViT, ConvNext)
• Implement search for adapted, layer-local objectives to train deep networks, based on decoder performance

Your Profile:

• Current master studies in biomedical engineering, physics, computer science, mathematics, electrical/electronic engineering or in a related field
• Strong programming skills (Python)
• Familiarity with machine learning and deep learning frameworks (e.g., PyTorch)
• Experience with HPC systems is a plus
• Ability to work independently and as part of a team
• Experimental enthusiasm is a must!

Please feel free to apply for the position even if you do not have all the required skills and knowledge. Missing skills can be learned.

Our Offer:

We work on the very latest issues that impact our society and are offering you the chance to actively help in shaping the change! We support you in your work with:

• An interesting and socially relevant topic for your thesis with future-oriented themes
• Ideal conditions for gaining practical experience alongside your studies
• An interdisciplinary collaboration on projects in an international, committed and collegial team
• Excellent technical equipment and the newest technology
• Qualified support through your scientific colleagues
• The chance to independently prepare and work on your tasks
• Flexible work (location) arrangements, e.g. remote work
• Flexible working hours as well as a reasonable remuneration

In addition to exciting tasks and a collaborative working atmosphere at Jülich, we have a lot more to offer: https://go.fzj.de/benefits

Place of employment: Aachen

We welcome applications from people with diverse backgrounds, e.g. in terms of age, gender, disability, sexual orientation / identity, and social, ethnic and religious origin. A diverse and inclusive working environment with equal opportunities in which everyone can realize their potential is important to us.