Abstract
We introduced the kernel trick to a linear generative model. In the present study, we trained a single layer model with nonnegativity constraint and expansive nonlinearity. After training, we found that the basis images acquired from natural scenes represented Gabor-like features. Moreover, the distributions of shape parameters of the basis images were similar to those found in V1. Other similar models, such as the sparse coding and the independent component analysis, fail to exhibit these properties.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
van Hateren, J.H., Ruderman, D.L.: Independent component analysis of natural image sequences yields spatiotemporal filters similar to simple cells in primary visual cortex. Proceeding of the Royal Society B: Biological Sciences 265(1412), 2315–2320 (1998)
Heeger, D.J.: Normalization of cell responses in cat striate cortex. Visual Neuroscience 9(2), 181–197 (1992)
Hoyer, P.O.: Non-negative matrix factorization with sparseness constraints. Journal of Machine Learning Research 5, 1457–1469 (2004)
Lee, D.D., Seung, H.S.: Algorithms for nonnegative matrix factorization. In: Advances in Neural Information Processing, ch. 13. MIT Press (2001)
Olshausen, B.A., Field, D.J.: Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381(6583), 607–609 (1996)
Ringach, D.L.: Spatial structure and symmetry of simple-cell receptive fields in macaque primary visual cortex. Journal of Neurophysiology 88(1), 455–463 (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yokoyama, H. (2013). Nonnegative Source Separation with Expansive Nonlinearity: Comparison with the Primary Visual Cortex. In: Lee, M., Hirose, A., Hou, ZG., Kil, R.M. (eds) Neural Information Processing. ICONIP 2013. Lecture Notes in Computer Science, vol 8227. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-42042-9_26
Download citation
DOI: https://doi.org/10.1007/978-3-642-42042-9_26
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-42041-2
Online ISBN: 978-3-642-42042-9
eBook Packages: Computer ScienceComputer Science (R0)