A Parameter-efficient Multi-subject Model for Predicting fMRI Activity
This work addresses the challenge of efficient fMRI prediction for neuroscience research, but it is incremental as it builds on existing pretrained models and linear encoding methods.
The authors tackled the problem of predicting fMRI activity across multiple subjects by introducing a parameter-efficient multi-subject model, achieving competitive results on the Algonauts 2023 benchmark.
This is the Algonauts 2023 submission report for team "BlobGPT". Our model consists of a multi-subject linear encoding head attached to a pretrained trunk model. The multi-subject head consists of three components: (1) a shared multi-layer feature projection, (2) shared plus subject-specific low-dimension linear transformations, and (3) a shared PCA fMRI embedding. In this report, we explain these components in more detail and present some experimental results. Our code is available at https://github.com/cmi-dair/algonauts23.