abstract = "The Donut problem requires separating two toroidal
distributions (classes) which are interlocked like
links in a chain. The cross-section of each
distribution is Gaussian distributed with standard
deviation sigma. This problem possesses a variety of
pathological traits: the mean of each distribution, for
example, lies in the densest point of the other.",
notes = "see also
http://www.cs.ucl.ac.uk/staff/W.Langdon/ftp/ftp.io.com/papers/.message
ICGA93.Donut.ps.Z - Preliminary version of Avi and
Walter's ICGA93 paper