Student projects in the EPFL NeuroAI Lab should:
- contribute to the mission of the lab, i.e. to create accurate models of behavior and underlying neural mechanisms — these models let us make sense of the mind and brain, and enable new applications in the diagnosis and treatment of neural disorders
- contribute to ongoing projects by permanent members (PhDs and Postdocs), so that the results of the project will be useful long-term, and so that the student can make use of existing expertise.
- be aligned to the interests of the student with a realistic timeline so that meaningful progress with some closure can be made over the duration of the project.
The list of projects here is non-comprehensive. It is always a good idea to reach out to lab members whose research you find interesting and inquire if you can contribute to their projects. In either case, we ask that you apply via the form on the lab website. During interviews, we will share a more comprehensive list of projects with you that we are thinking about.
Testing computer vision models on human illusions
(1 student)
In a collaborative project with Michael Herzog’s group (Life science), we are looking for a Master student with strong computational skills and an interest in comparing models to human behavior. The goal of the project is to compare models from computer vision (including CNNs and vision transformers) with human performance in visual tasks with a focus on visual illusions (human data are already collected). We will build human behavioral benchmarks on Brain-Score to evaluate a wide range of models on their alignment to humans. This is a larger effort and we will prioritize students who are able to commit for at least 6 months.
References:
Advancing Benchmarks and Models for Dynamic Vision in the Brain
(1~2 students)
Project lead: Yingtian (David) Tang
Project description:
The project aims to build state-of-the-art benchmarks and models of the dynamic visual processing in the human brain. It will build upon an existing benchmark framework (https://dynamic-vision.epfl.ch/) and explore upcoming neural recordings (like fMRI on video watching), foundation models (like VJEPAs and large VLMs), as well as the most efficient way to build model-brain decoders.
The project expects students with strong engineering skills, especially in deep learning (PyTorch). Additional experience on neural data (fMRI) processing is also highly recommended. The project suits students with a goal of building engineer background (as the outcomes go to the open-source Brain-Score platform https://github.com/brain-score/vision) and exploring the field of NeuroAI. The project outcomes could also be potentially organized as publications.
References:
- https://www.biorxiv.org/content/10.1101/2025.07.22.664908v1
- https://www.biorxiv.org/content/10.1101/407007v1
- https://arxiv.org/html/2508.10784v1
Building Brain-like Computation Topography
(1 student)
Project lead: Yingtian (David) Tang
Project description:
The project aims to build a Transformer model in which the “computational wiring” (ie, the attention) closely resembles that in the human brain (ie, as reflected by the functional organization). It builds on the observation of existing low-level topographical matching between the model and the brain (retinotopic regions) and tries to advance to higher-level regions, by carefully designing a hierarchy of “registers”.
The project expects students with very strong skills in deep learning (ViT architectures, SOTA representation learning like MAE, etc) and a certain level of independent research. Additional experience on human task fMRI is also highly recommended. The project suits students with a goal of conducting advanced NeuroAI research. The project targets publications in top machine learning conferences and neuroscience journals.
References: