Editing Random forest (section)

=== History ===
[[Leo Breiman]]<ref name="breiman2000some">{{cite journal | first = Leo | last = Breiman | author-link = Leo Breiman | title = Some infinity theory for predictor ensembles | institution = Technical Report 579, Statistics Dept. UCB | year = 2000 | url = https://statistics.berkeley.edu/tech-reports/579 }}</ref> was the first person to notice the link between random forest and [[kernel methods]]. He pointed out that random forests trained using [[i.i.d.]] random vectors in the tree construction are equivalent to a kernel acting on the true margin. Lin and Jeon<ref name="lin2006random">{{cite journal | first1 = Yi | last1 = Lin | first2 = Yongho | last2 = Jeon | title = Random forests and adaptive nearest neighbors | journal = Journal of the American Statistical Association | volume = 101 | number = 474 | pages = 578–590 | year = 2006 | doi = 10.1198/016214505000001230 | citeseerx = 10.1.1.153.9168 | s2cid = 2469856 }}</ref> established the connection between random forests and adaptive nearest neighbor, implying that random forests can be seen as adaptive kernel estimates. Davies and Ghahramani<ref name="davies2014random">{{cite arXiv |first1=Alex |last1=Davies |first2=Zoubin|last2=Ghahramani |title=The Random Forest Kernel and other kernels for big data from random partitions |eprint=1402.4293 |year= 2014 |class=stat.ML }}</ref> proposed Kernel Random Forest (KeRF) and showed that it can empirically outperform state-of-art kernel methods. Scornet<ref name="scornet2015random"/> first defined KeRF estimates and gave the explicit link between KeRF estimates and random forest. He also gave explicit expressions for kernels based on centered random forest<ref name="breiman2004consistency">{{cite journal | first1 = Leo | last1 = Breiman | first2 = Zoubin | last2 = Ghahramani | name-list-style = vanc | title = Consistency for a simple model of random forests | journal = Statistical Department, University of California at Berkeley. Technical Report | number = 670 | year = 2004 | citeseerx = 10.1.1.618.90 }}</ref> and uniform random forest,<ref name="arlot2014analysis">{{cite arXiv |first1=Sylvain |last1=Arlot  | first2 = Robin | last2 = Genuer | name-list-style = vanc |title=Analysis of purely random forests bias |eprint=1407.3939 |year= 2014 |class=math.ST  }}</ref> two simplified models of random forest. He named these two KeRFs Centered KeRF and Uniform KeRF, and proved upper bounds on their rates of consistency.