Deep unsupervised learning on a desktop PC a primer for connitive scientist外文电子书籍.pdf

METHODS ARTICLE published: 06 May 2013 doi: 10.3389/fpsyg.2013.00251 Deep unsupervised learning on a desktop PC: a primer for cognitive scientists 1 1 1 1,2 Alberto Testolin *, Ivilin Stoianov , Michele De Filippo De Grazia and Marco Zorzi * 1 Computational Cognitive Neuroscience Lab, Department of General Psychology, University of Padova, Padova, Italy 2 IRCCS San Camillo Neurorehabilitation Hospital, Venice Lido, Italy Edited by: Deep belief networks hold great promise for the simulation of human cognition because Christoph T. Weidemann, Swansea they show how structured and abstract representations may emerge from probabilistic University, UK unsupervised learning. These networks build a hierarchy of progressively more complex Reviewed by: Bradley Love, University College distributed representations of the sensory data by fitting a hierarchical generative model. London, UK However, learning in deep networks typically requires big datasets and it can involve mil- Simon Farrell, University of Bristol, UK lions of connection weights, which implies that simulations on standard computers are *Correspondence: unfeasible. Developing realistic, medium-to-large-scale learning models of cognition would Alberto Testolin and Marco Zorzi, therefore seem to require expertise in programing