Approximate the Natural Neighbors...

[aupetit]

Hi,

I am working in the field of artificial Neural Networks.
I'd like to use Natural Neighbors interpolation techniques
in that field but I come up against a difficulty: the dimension
of the data space is quite high (for example 10 dimensions)
and so it seems that NN are not a feasible technique in such
a case.
Does it exist a method to approximate the natural neighbors
and eventually to approximate the natural neighbors coordinates
in the case of high-dimensional spaces?

Thank you

Best regards
Mishaël Aupetit
PhD Student in Neural Networks
LGI2P-EMA, France



-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request at research.bell-labs.com with the line:
send readme
Now archived at http://uiuc.edu/~sariel/CG/compgeom/maillist.html.