My Projects

• Neural Unbalanced Optimal Transport for Modeling Single-Cell Dynamics

  Harvard & ETH, December 2022

  Spotlight Presentation at NeurIPS Workshop 'Learning Meaningful Representations of Life'

  Tracking the development of cells over time is a major challenge in biology, as measuring cells usually requires their destruction. Optimal transport (OT) can help solve this challenge by learning an optimal coupling of samples taken at different points in time. In this work, we propose a novel framework for unbalanced OT.

• Self-Supervised Learning on Protein Structures based on Geometric Deep Learning

  ETH, February 2022

  Learning on 3D protein structures is an emerging area of machine learning research with promising applications in drug discovery and material innovation. In this semester project at ETH, we explored pre-training strategies for protein structures using Graph Neural Networks.