| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
l.wiskott{at}biologie.hu-berlin.de, Computational Neurobiology Laboratory, Salk Institute for Biological Studies, San Diego, CA 92168, U.S.A.; Institute for Advanced Studies, D-14193, Berlin, Germany; and Innovationskolleg Theoretische Biologie, Institute for Biology, Humboldt-University Berlin, D-10115 Berlin, Germany
terry{at}salk.edu, Howard Hughes Medical Institute, The Salk Institute for Biological Studies, La Jolla, CA 92037, U.S.A., and Department of Biology, University of California at San Diego, La Jolla, CA 92037, U.S.A.
Invariant features of temporally varying signals are useful for analysis and classification. Slow feature analysis (SFA) is a new method for learning invariant or slowly varying features from a vectorial input signal. It is based on a nonlinear expansion of the input signal and application of principal component analysis to this expanded signal and its time derivative. It is guaranteed to find the optimal solution within a family of functions directly and can learn to extract a large number of decorrelated features, which are ordered by their degree of invariance. SFA can be applied hierarchically to process high-dimensional input signals and extract complex features. SFA is applied first to complex cell tuning properties based on simple cell output, including disparity and motion. Then more complicated input-output functions are learned by repeated application of SFA. Finally, a hierarchical network of SFA modules is presented as a simple model of the visual system. The same unstructured network can learn translation, size, rotation, contrast, or, to a lesser degree, illumination invariance for one-dimensional objects, depending on only the training stimulus. Surprisingly, only a few training objects suffice to achieve good generalization to new objects. The generated representation is suitable for object recognition. Performance degrades if the network is trained to learn multiple invariances simultaneously.
This article has been cited by other articles:
|
|
 |
|
  D. D. Cox and J. J. DiCarlo Does Learned Shape Selectivity in Inferior Temporal Cortex Automatically Generalize Across Retinal Position? J. Neurosci., October 1, 2008; 28(40): 10045 - 10055. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Li and J. J. DiCarlo Unsupervised Natural Experience Rapidly Alters Invariant Object Representation in Visual Cortex Science, September 12, 2008; 321(5895): 1502 - 1507. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Byrne and S. Becker A principle for learning egocentric-allocentric transformation. Neural Comput., March 1, 2008; 20(3): 709 - 737. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Miao and R. P. N. Rao Learning the Lie Groups of Visual Invariance Neural Comput., October 1, 2007; 19(10): 2665 - 2693. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Cadieu, M. Kouh, A. Pasupathy, C. E. Connor, M. Riesenhuber, and T. Poggio A Model of V4 Shape Selectivity and Invariance J Neurophysiol, September 1, 2007; 98(3): 1733 - 1750. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Blaschke, T. Zito, and L. Wiskott Independent slow feature analysis and nonlinear blind source separation. Neural Comput., April 1, 2007; 19(4): 994 - 1021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Turner and M. Sahani A maximum-likelihood interpretation for slow feature analysis. Neural Comput., April 1, 2007; 19(4): 1022 - 1038. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Creutzig and H. Sprekeler Predictive Coding and the Slowness Principle: An Information-Theoretic Approach Neural Comput., April 1, 2007; 20(4): 1026 - 1041. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Schwartz, T. J. Sejnowski, and P. Dayan Soft mixer assignment in a hierarchical generative model of natural scene statistics. Neural Comput., November 1, 2006; 18(11): 2680 - 2718. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Blaschke, P. Berkes, and L. Wiskott What Is the Relation Between Slow Feature Analysis and Independent Component Analysis? Neural Comput., October 1, 2006; 18(10): 2495 - 2508. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Berkes and L. Wiskott On the Analysis and Interpretation of Inhomogeneous Quadratic Forms as Receptive Fields Neural Comput., August 1, 2006; 18(8): 1868 - 1895. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Karklin and M. S. Lewicki A Hierarchical Bayesian Model for Learning Nonlinear Statistical Regularities in Nonstationary Natural Signals Neural Comput., February 1, 2005; 17(2): 397 - 423. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. B. Grimes and R. P. N. Rao Bilinear Sparse Coding for Invariant Vision Neural Comput., January 1, 2005; 17(1): 47 - 73. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. P. Kording, C. Kayser, W. Einhauser, and P. Konig How Are Complex Cell Properties Adapted to the Statistics of Natural Stimuli? J Neurophysiol, January 1, 2004; 91(1): 206 - 212. [Abstract] [Full Text]
|
|
|
|
|
|
L. Wiskott Slow Feature Analysis: A Theoretical Analysis of Optimal Free Responses Neural Comput., September 1, 2003; 15(9): 2147 - 2177. [Abstract] [Full Text]
|
|
|
|
|
|
C. Kayser, K. P. Kording, and P. Konig Learning the Nonlinearity of Neurons from Natural Visual Stimuli Neural Comput., August 1, 2003; 15(8): 1751 - 1759. [Abstract] [Full Text]
|
|
|
|
|
|
H. Wersing and E. Korner Learning Optimized Features for Hierarchical Models of Invariant Object Recognition Neural Comput., July 1, 2003; 15(7): 1559 - 1588. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Hurri and A. Hyvarinen Simple-Cell-Like Receptive Fields Maximize Temporal Coherence in Natural Video Neural Comput., March 1, 2003; 15(3): 663 - 691. [Abstract] [Full Text] [PDF]
|
|
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| J COGNITIVE NEUROSCIENCE | NEURAL COMPUTATION | MIT PRESS JOURNALS |
