Interpreting uninterpretable predictors:kernel methods,Shtarkov solutions,and random forests  

作　　者：T.M.Le Bertrand Clarke 

机构地区：[1]Department of mathematics,science,and informatics,mercer university,atlanta,ga,usa [2]department of statistics,university of nebraska-lincoln,lincoln,ne,usa

出　　处：《Statistical Theory and Related Fields》2022年第1期10-28,共19页统计理论及其应用（英文）

摘　　要：Many of the best predictors for complex problems are typically regarded as hard to interpret physically.These include kernel methods,Shtarkov solutions,and random forests.We show that,despite the inability to interpret these three predictors to infinite precision,they can be asymptotically approximated and admit conceptual interpretations in terms of their mathe-matical/statistical properties.The resulting expressions can be in terms of polynomials,basis elements,or other functions that an analyst may regard as interpretable.

关 键 词：BAYES BOOSTING kernel methods random forest Shtarkov predictor STACKING 

分 类 号：O17[理学—数学]