distribution used in adaboost

[ Follow Ups ] [ Post Followup ] [ Message Board on Boosting and Related Methods ] [ FAQ ]

Posted by Bart Hamers on June 07, 19102 at 18:11:54:

If one uses a a distribution to choose your training points, does it make a difference if this is done with or without replacement?
Without replacement gives an extra weigth to your trainingpoint in a next iteration. Does this disturb the boosting weigths in adaboost?

Bart

Follow Ups:

Post a Followup

Name:
E-Mail:
Subject:
Comments:	: If one uses a a distribution to choose your training points, does it make a difference if this is done with or without replacement? : Without replacement gives an extra weigth to your trainingpoint in a next iteration. Does this disturb the boosting weigths in adaboost? : Bart
Optional Link (URL):
Link Title:
Optional Image (URL):