Jump to content

Factorial code

From Wikipedia, the free encyclopedia
This is an old revision of this page, as edited by Algorithms (talk | contribs) at 19:07, 5 June 2007 (a central concept of unsupervised learning). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.
(diff) ← Previous revision | Latest revision (diff) | Newer revision → (diff)

Most real world data sets consist of data vectors whose individual components are not statistically independent, that is, they are redundant in the statistical sense. Then it is desirable to create a factorial code of the data, i. e., a new vector-valued representation of each data vector such that each original data is uniquely encoded by the corresponding code vector (loss-free coding), but the code components are statistically independent.