Hierarchical K-Means Algorithm for Modeling Visual Area V2 Neurons

Computational studies about the properties of the receptive fields of neurons in the cortical visual pathway of mammals are abundant in the literature but most addressed neurons in the primary visual area (V1). Recently, the sparse deep belief network (DBN) was proposed to model the response properties of neurons in the V2 area. By investigating the factors that contribute to the success of the model, we find that a simple algorithm for data clustering, K-means algorithm can be stacked into a hierarchy to reproduce these properties of V2 neurons, too. In addition, it is computationally much more efficient than the sparse DBN.

[1]  David J. Field,et al.  Sparse coding with an overcomplete basis set: A strategy employed by V1? , 1997, Vision Research.

[2]  Geoffrey E. Hinton.  Training Products of Experts by Minimizing Contrastive Divergence , 2002, Neural Computation.

[3]  T. Poggio,et al.  Hierarchical models of object recognition in cortex , 1999, Nature Neuroscience.

[4]  David J. Field,et al.  Emergence of simple-cell receptive field properties by learning a sparse code for natural images , 1996, Nature.

[5]  Marc'Aurelio Ranzato,et al.  Sparse Feature Learning for Deep Belief Networks , 2007, NIPS.

[6]  Honglak Lee,et al.  Sparse deep belief net model for visual area V2 , 2007, NIPS.

[7]  Honglak Lee,et al.  Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations , 2009, ICML '09.

[8]  Andrew Y. Ng,et al.  Unsupervised learning models of primary cortical receptive fields and receptive field plasticity , 2011, NIPS.

[9]  T. Poggio,et al.  A model of V4 shape selectivity and invariance. , 2007, Journal of neurophysiology.

[10]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[11]  Minami Ito,et al.  Representation of Angles Embedded within Contour Stimuli in Area V2 of Macaque Monkeys , 2004, The Journal of Neuroscience.

[12]  Michael S. Lewicki,et al.  A Hierarchical Bayesian Model for Learning Nonlinear Statistical Regularities in Nonstationary Natural Signals , 2005, Neural Computation.

[13]  Terrence J. Sejnowski,et al.  The “independent components” of natural scenes are edge filters , 1997, Vision Research.

[14]  Michael S. Lewicki,et al.  Emergence of complex cell properties by learning to generalize in natural scenes , 2009, Nature.

[15]  D H HUBEL,et al.  RECEPTIVE FIELDS AND FUNCTIONAL ARCHITECTURE IN TWO NONSTRIATE VISUAL AREAS (18 AND 19) OF THE CAT. , 1965, Journal of neurophysiology.

[16]  Honglak Lee,et al.  An Analysis of Single-Layer Networks in Unsupervised Feature Learning , 2011, AISTATS.