Student Research Assistant Position

Project description

A major limitation of deep learning for medical applications is the scarcity of labelled data. Meta-learning, which leverages principles learned from previous tasks for new tasks, has the potential to mitigate this data scarcity. However, most meta-learning methods assume idealised settings with homogeneous task definitions. The most widely used family of meta-learning methods, those based on Model-Agnostic Meta-Learning (MAML), require a constant network architecture and therefore a fixed number of classes per classification task.

We take a first step in the direction of making meta-learning algorithms, suitable for more realistic medical problems by investigating different strategies for training and testing with a variable number of possible labels.

To this end, we are currently assembling a dataset and writing a PyTorch toolbox for testing meta-learning algorithms in a realistic medical setting. We aim to make our dataset and toolbox publicly available and to issue a challenge for the MICCAI 2023 conference.

For more information visit our project website or watch the video abstract of our past paper.

Your tasks

You will aid in assembling the dataset for the challenge and writing the PyTorch medical meta-learning toolbox. Your main tasks will include collecting meta-information in a structured way for multiple source datasets and writing PyTorch dataset classes facilitating meta-learning on these datasets.

Your profile

• Good knowledge of Machine Learning and of Image Analysis/Computer Vision,
• Interest in working with medical imaging datasets,
• Interest in learning about and working on meta-learning,
• Proficiency in Python, PyTorch and Git.

What we offer

• HiWi salary according to the standard rates of the University of Tübingen,
• A desk space in the Tübingen AI Research Building,
• Insights into an exciting and trending research field,
• The possibility of contributing to a scientific publication.

How to apply

If interested, please contact Stefano Woerner and attach your CV and transcript of records to apply. Only currently enrolled students of the University of Tübingen can be considered for this position.