papers on ICA

[Aapo Hyvarinen]

Dear Connectionists,

the following papers on extensions of ICA can now be found on
my web page. 

- Aapo Hyvarinen
  http://www.cis.hut.fi/~aapo/

---------------------------------------------------------------------

A. Hyvarinen and P. Hoyer.
Topographic Independent Component Analysis.

http://www.cis.hut.fi/~aapo/ps/gz/TICA.ps.gz

Abstract: In ordinary independent component analysis, the components
are assumed to be completely independent, and they do not necessarily
have any meaningful order relationships. In practice, however, the
estimated ``independent'' components are often not at all
independent. We propose that this residual dependence structure could
be used to define a topographic order for the components. In
particular, a distance between two components could be defined using
their higher-order correlations, and this distance could be used to
create a topographic representation. Thus we obtain a linear
decomposition into approximately independent components, where the
dependence of two components is approximated by the proximity of the
components in the topographic representation.

---------------------------------------------------------------------

A. Hyvarinen and P. Hoyer.
Emergence of phase and shift invariant features
by decomposition of natural images into independent feature subspaces.
(to appear in Neural Computation)

http://www.cis.hut.fi/~aapo/ps/gz/NC99_complex.ps.gz

Olshausen and Field (1996) applied the principle of independence
maximization by sparse coding to extract features from natural
images. This leads to the emergence of oriented linear filters that
have simultaneous localization in space and in frequency, thus
resembling Gabor functions and simple cell receptive fields.  In this
paper, we show that the same principle of independence maximization
can explain the emergence of phase and shift invariant features,
similar to those found in complex cells. This new kind of emergence is
obtained by maximizing the independence between norms of projections
on linear subspaces (instead of the independence of simple linear
filter outputs). The norms of the projections on such `independent
feature subspaces' then indicate the values of invariant features.

----------------------------------------------------------------------

(Some other new papers on ICA can be found on my publication page
http://www.cis.hut.fi/~aapo/pub.html   as well.)