Asymptotic Distributions and Rates of Convergence for Random Forests via Generalized U-statistics

Abstract

Random forests remain among the most popular off-the-shelf supervised learning algorithms. Despite their well-documented empirical success, however, until recently, few theoretical results were available to describe their performance and behavior. In this work we push beyond recent work on consistency and asymptotic normality by establishing rates of convergence for random forests and other supervised learning ensembles. We develop the notion of generalized U-statistics and show that within this framework, random forest predictions can potentially remain asymptotically normal for larger subsample sizes than previously established. We also provide Berry-Esseen bounds in order to quantify the rate at which this convergence occurs, making explicit the roles of the subsample size and the number of trees in determining the distribution of random forest predictions.

76 pages, 7 figure

References in corpus (5)

• Normal approximation under local dependence
• Approximating high-dimensional infinite-order $U$-statistics: statistical and computational guarantees
• Making Sense of Random Forest Probabilities: a Kernel Perspective
• Scalable and Efficient Hypothesis Testing with Random Forests
• Getting Better from Worse: Augmented Bagging and a Cautionary Tale of Variable Importance

Cited by in corpus (8)

• Randomization as Regularization: A Degrees of Freedom Explanation for Random Forest Success
• Getting Better from Worse: Augmented Bagging and a Cautionary Tale of Variable Importance
• $V$-statistics and Variance Estimation
• Bias, Consistency, and Alternative Perspectives of the Infinitesimal Jackknife
• Forest Guided Smoothing
• On the Subbagging Estimation for Massive Data
• Random Planted Forest: a directly interpretable tree ensemble
• Locally Optimized Random Forests