Scaling up ridge regression for brain encoding in a massive individual fMRI dataset

arXiv:2403.19421v1 Announce Type: new
Abstract: Brain encoding with neuroimaging data is an established analysis aimed at predicting human brain activity directly from complex stimuli features such as movie frames. Typically, these features are the latent space representation from an artificial neural network, and the stimuli are image, audio, or text inputs. Ridge regression is a popular prediction model for brain encoding due to its good out-of-sample generalization performance. However, training a ridge regression model can be highly time-consuming when dealing …

abstract analysis artificial arxiv brain brain activity cs.ai cs.lg data dataset encoding features fmri human image massive movie movıe network neural network neuroimaging q-bio.nc q-bio.qm regression representation ridge scaling scaling up space type