Curse Of Dimensionality

Curse of dimensionality is a term coined by Richard Bellman applied to the problem caused by the rapid increase in volume associated with adding extra dimensions to a (mathematical) space. Breiman gives as an example the fact that 100 observations cover the one-dimensional unit interval 0,1 on the real line quite well. One could draw a histogram of the results, and draw inferences. If one now considers the corresponding 10-dimensional unit hypersquare, 100 observations are now isolated points in a vast empty space. To get similar coverage to the one-dimensional space would now require 1020 observations, which is at least a massive undertaking and may well be impractical. The curse of dimensionality is a significant obstacle in machine learning problems that involve learning from few data samples in a high-dimensional feature space. See also: quasi-random.

Reference

Bellman, R.E. 1961. Adaptive Control Processes. Princeton University Press, Princeton, NJ.

 

<< PreviousWord BrowserNext >>
kalmar nyckel
schwarzbier
soap plant
alfred stelzner
torrance high school
stefan panaretov
accademia dell' arte del disegno
arsenic and old lace (movie)
granada hills, los angeles, california
gnther von kluge
calaveras
international rules football
public holidays in qatar
wilhelm list
karel dillen
just one of the guys (full house episode)
winesburg
backcrossing
frequency list
captain tugg
hugo sperrle
ilium (kurt vonnegut)
unifil
nathaniel crisp
hungarian democratic forum
paul ludwig ewald von kleist
doorknob
gush katif
queens domain
pagiran muda mohamed bolkiah
hamad ibn jaber al thani
eduard von bhm ermolli
sugar house (salt lake city)
leroy bach
socialist party new psi
birmingham gun quarter
ertl company
jay bennett
pao fa temple
prince sa'ud al faysal
kwanza
klaus roth
reynaldo hahn
gledhill