Timezone: »

Semi-Supervised Factored Logistic Regression for High-Dimensional Neuroimaging Data
Danilo Bzdok · Michael Eickenberg · Olivier Grisel · Bertrand Thirion · Gael Varoquaux

Mon Dec 07 04:00 PM -- 08:59 PM (PST) @ 210 C #14

Imaging neuroscience links human behavior to aspects of brain biology in ever-increasing datasets. Existing neuroimaging methods typically perform either discovery of unknown neural structure or testing of neural structure associated with mental tasks. However, testing hypotheses on the neural correlates underlying larger sets of mental tasks necessitates adequate representations for the observations. We therefore propose to blend representation modelling and task classification into a unified statistical learning problem. A multinomial logistic regression is introduced that is constrained by factored coefficients and coupled with an autoencoder. We show that this approach yields more accurate and interpretable neural models of psychological tasks in a reference dataset, as well as better generalization to other datasets.

Author Information

Danilo Bzdok (INRIA)
Michael Eickenberg (Flatiron Institute)
Olivier Grisel (Inria)
Bertrand Thirion (INRIA)
Gael Varoquaux (Parietal Team, INRIA)

More from the Same Authors