Neural decoding with hierarchical generative models (2010)

Titel Neural decoding with hierarchical generative models
Gepubliceerd in Neural Computation, Vol. 22, p.3127-3142. ISSN 0899-7667.
Auteur Gerven, M.A.J. van; Lange, F.P. de; Heskes, T.M.
Datum 2010
Type Artikel
Samenvatting Recent research has shown that reconstruction of perceived images based on hemodynamic response as measured with functional magnetic resonance imaging (fMRI) is starting to become feasible. In this letter, we explore reconstruction based on a learned hierarchy of features by employing a hierarchical generative model that consists of conditional restricted Boltzmann machines. In an unsupervised phase, we learn a hierarchy of features from data, and in a supervised phase, we learn how brain activity predicts the states of those features. Reconstruction is achieved by sampling from the model, conditioned on brain activity. We show that by using the hierarchical generative model, we can obtain good-quality reconstructions of visual images of handwritten digits presented during an fMRI scanning session.
OpenURL Zoek deze publicatie in (uw) bibliotheek
Persistent Identifier urn:nbn:nl:ui:22-2066/90034
Metadata XML
Bron Radboud Universiteit Nijmegen

Ga terug naar de inhoud
Ga terug naar de site navigatie