Building a foundation model for the brain

datasets, theory, and models

Workshop

Advances in neurotechnology have enabled the collection of large-scale neural recordings during animal behavior. To extract insights from these datasets, researchers are now faced with significant challenges when comparing data across different brain regions, subjects, and tasks. Inspired by recent successes in large-scale "foundation" modeling of natural language and computational biology, there has been a push towards developing "neuro-foundation models" that can be trained across diverse datasets and then fine-tuned for downstream tasks and analyses including neural encoding, decoding, cell-type classification, and activity prediction. In this workshop, we aim to provide an overview of current neuro-foundation model approaches and to bring together experimentalists, theoreticians, and model builders to discuss how this paradigm can lead to a deeper understanding of the brain. In particular, we plan to highlight the utility of neuro-foundation models in analyses across recording modalities, brain regions, individuals, species, and contexts. In addition to these talks, we will also host a panel to discuss challenges, opportunities, and risks of building these models.

This workshop will bring together experts who are at the forefront of neuro-foundation model research. Our goal is to foster a much-needed dialogue between modelers, theorists, and experimentalists, providing a platform to explore the unique challenges, opportunities, and risks of creating foundation models for brain data. Participants will discuss the future of large-scale neural modeling through diverse perspectives, including:

• Cross-region: Exploring how neural activity across different brain regions and cell types can be jointly analyzed.
• Cross-animal and cross-species: Investigating common and unique neural signatures across individuals and species.
• Cross-task and cross-context: Understanding how neural activity adapts across different behaviors and environments.

In addition to these talks, we will host a panel discussion focused on the current landscape and future directions of neuro-foundation models. This discussion will delve into critical topics such as:

• The potential applications of foundation models for neural data
• Key challenges in developing generalizable models across diverse datasets
• Ethical considerations and risks of scaling these models

By bringing together diverse perspectives, this workshop will serve as a catalyst for identifying commonalities and differences in current approaches, highlighting how these models can generate new insights into brain function, and establishing a roadmap for this growing interdisciplinary field.