Deep probabilistic model synthesis enables unified modeling of whole-brain neural activity across individual subjects
This addresses the need for unified modeling across individual subjects in neuroscience, offering a novel framework for data synthesis.
The authors tackled the problem of synthesizing quantitative models across multiple system instances, such as combining neural data from individual animals, by introducing deep probabilistic model synthesis (DPMS), which improved upon single-instance models on synthetic and whole-brain neural activity data from larval zebrafish.
Many disciplines need quantitative models that synthesize experimental data across multiple instances of the same general system. For example, neuroscientists must combine data from the brains of many individual animals to understand the species' brain in general. However, typical machine learning models treat one system instance at a time. Here we introduce a machine learning framework, deep probabilistic model synthesis (DPMS), that leverages system properties auxiliary to the model to combine data across system instances. DPMS specifically uses variational inference to learn a conditional prior distribution and instance-specific posterior distributions over model parameters that respectively tie together the system instances and capture their unique structure. DPMS can synthesize a wide variety of model classes, such as those for regression, classification, and dimensionality reduction, and we demonstrate its ability to improve upon single-instance models on synthetic data and whole-brain neural activity data from larval zebrafish.