Learning representations in Bayesian Confidence Propagation neural networks
作者： Pawel Herman , Anders Lansner , Naresh Balaji Ravichandran
DOI:
关键词:
摘要: Unsupervised learning of hierarchical representations has been one the most vibrant research directions in deep during recent years. In this work we study biologically inspired unsupervised strategies neural networks based on local Hebbian learning. We propose new mechanisms to extend Bayesian Confidence Propagating Neural Network (BCPNN) architecture, and demonstrate their capability for salient hidden when tested MNIST dataset.
arxiv.org
harvard.edu参考文章(18)
Anders Lansner, Simon Benjaminsson, Christopher Johansson, From ANN to Biomimetic Information Processing OSPEL Workshop on Bio-inspired Signal Processing Barcelona, SPAIN, 2007. ,vol. 188, pp. 33- 43 ,(2009) , 10.1007/978-3-642-00176-5_2
Pierre Berthet, Jeanette Hellgren-Kotaleski, Anders Lansner, Action selection performance of a reconfigurable basal ganglia inspired model with Hebbian-Bayesian Go-NoGo connectivity. Frontiers in Behavioral Neuroscience. ,vol. 6, pp. 65- 65 ,(2012) , 10.3389/FNBEH.2012.00065
Rodney J. Douglas, Kevan A.C. Martin, Recurrent neuronal circuits in the neocortex. Current Biology. ,vol. 17, pp. 496- 500 ,(2007) , 10.1016/J.CUB.2007.04.024
Markus Butz, Florentin Wörgötter, Arjen van Ooyen, Activity-dependent structural plasticity Brain Research Reviews. ,vol. 60, pp. 287- 305 ,(2009) , 10.1016/J.BRAINRESREV.2008.12.023
Rockland, Five Points on Columns Frontiers in Neuroanatomy. ,vol. 4, pp. 22- 22 ,(2010) , 10.3389/FNANA.2010.00022
V. Mountcastle, The columnar organization of the neocortex. Brain. ,vol. 120, pp. 701- 722 ,(1997) , 10.1093/BRAIN/120.4.701
Anders Lansner, Associative memory models: from the cell-assembly theory to biophysically detailed cortex simulations. Trends in Neurosciences. ,vol. 32, pp. 178- 186 ,(2009) , 10.1016/J.TINS.2008.12.002
Philip J. Tully, Matthias H. Hennig, Anders Lansner, Synaptic and nonsynaptic plasticity approximating probabilistic inference. Frontiers in Synaptic Neuroscience. ,vol. 6, pp. 8- 8 ,(2014) , 10.3389/FNSYN.2014.00008
Yoshua Bengio, Pierre-Antoine Manzagol, Samy Bengio, Dumitru Erhan, Aaron Courville, Pascal Vincent, Why Does Unsupervised Pre-training Help Deep Learning? Journal of Machine Learning Research. ,vol. 11, pp. 625- 660 ,(2010)