When building an ensemble model (eg random forest), are there good ways to group observations known to be highly correlated rather than IID?
-
-
Replying to @patrickc
Seems it should increase the bias of the individual weak classifiers but avoid systematic bias across the ensemble. (cc
@posco,@avibryant.)1 reply 0 retweets 2 likes -
Replying to @patrickc
@patrickc@posco@avibryant http://www.informatik.uni-ulm.de/ni/staff/HKestler/Reisensburg2009/talks/adler.pdf …2 replies 1 retweet 3 likes
Replying to @josh_wills
@josh_wills @posco @avibryant Awesome, thanks.
4:16 PM - 8 Dec 2014
0 replies
0 retweets
0 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.