Master’s thesis , Optimal transport for perturbation modeling

Master’s thesis

Optimal transport for perturbation modeling

Ref. 2023_006

A master’s thesis student position is available for a highly interdisciplinary project at the intersection of machine learning, quantum computing and single-cell biology.

To explore novel and impactful applications of quantum computing in healthcare and drug discovery, this project plans to build upon recent advances in single-cell methods which allow to capture cellular and metabolic activity with unprecedented spatiotemporal resolution and throughput.

The overall goal of this thesis is to devise a novel conditional Optimal Transport (OT) approach coupled with machine learning (e.g., input-convex neural networks) to model perturbational effects of therapies at a cell-level resolution. After benchmarking state-of-the-art approaches (e.g., variational autoencoders [1], neural OT [2]) on existing datasets on single and combinatorial perturbations, the candidate will also experiment with Quantum Machine Learning approaches able to learn the perturbation-based transportation plan that maps cells from before to after the perturbations.

The successful candidate will join the AI for Scientific Discovery group at IBM Research Europe in Zurich and will strongly collaborate with Quantum computing researchers across four IBM Research labs to make the developed algorithms amenable to execution on real quantum hardware.

The project is available for a minimum duration of six months. Please note that this is a non-remunerated M.Sc. thesis project.

Qualifications

• Enrolled in a masters degree, preferably in data or computer science, applied mathematics, bio-/chemoinformatics or engineering
• Strong theoretical background in mathematical optimization and machine learning is highly desired
• Familiarity with quantum computing fundamentals is a strong plus
• An excellent academic record
• Strong programming skills in Python
• Familiarity with Linux environments and software development tools (git/GitHub, IDEs, etc.)
• High amount of creativity and outstanding problem-solving ability

Diversity

IBM is committed to diversity at the workplace. With us you will find an open, multicultural environment. Excellent flexible working arrangements enable all genders to strike the desired balance between their professional development and their personal lives.

How to apply

Please submit your CV including contact information for two or three references. We encourage candidates to also share a 3-minute video, in which they introduce themselves, as well as highlight their motivation and expertise. The video is not mandatory.

References
[1] Lotfollahi, M., Wolf, F.A. & Theis, F.J. scGen predicts single-cell perturbation responses. Nat Methods 16, 715-721 (2019).
[2] Bunne, C., Krause, A., & Cuturi, M. (2022). Supervised training of conditional monge maps. NeurIPS 2022.