arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2306.04423cs.LGcs.DS
收藏

关于计算最优树集成

On Computing Optimal Tree Ensembles

TU Berlin(柏林工业大学)

Christian Komusiewicz, Pascal Kunz, Frank Sommer, Manuel Sorge

AI总结:

本文针对树集成缺乏最优计算研究的问题,提出了两种新算法及下界:利用见证树技术计算指定大小内最优树集成,并证明动态规划适用于树集成,同时发现集成所需切割数随树增加呈指数级减少。

摘要:

随机森林以及更一般的(决策)树集成是广泛用于分类和回归的方法。近期的算法进展使得计算在各种度量(如其大小或深度)下最优的决策树成为可能。我们尚未发现针对树集成的此类研究,并旨在为该领域做出贡献。我们主要提供了两种新算法及相应的下界。首先,我们能够延续并大幅改进决策树的可处理性结果:我们提出了一种算法,给定一个训练数据集和一个大小界限 $S \in \mathbb{R}$,能够计算出一个大小至多为 $S$ 且能正确分类数据的树集成。该算法的运行时间为 $(4δD S)^S \cdot poly$,其中 $D$ 是最大定义域大小,$δ$ 是两个样本不同的最大特征数,$n$ 是输入样本数,$poly$ 是输入大小的多项式。对于决策树,即大小为 1 的集成,我们得到的运行时间为 $(δD s)^s \cdot poly$,其中 $s$ 是树的大小。为了获得这些算法,我们引入了见证树技术,该技术在实际实现中看起来很有前景。其次,我们证明了已成功应用于计算决策树的动态规划也可能适用于树集成,提供了一种运行时间为 $\ell^n \cdot poly$ 的算法,其中 $\ell$ 是树的数量。最后,我们比较了决策树和树集成对训练数据集进行分类所需的切割数量,表明随着树数量的增加,集成可能需要呈指数级减少的切割次数。

英文摘要:

Random forests and, more generally, (decision\nobreakdash-)tree ensembles are widely used methods for classification and regression. Recent algorithmic advances allow to compute decision trees that are optimal for various measures such as their size or depth. We are not aware of such research for tree ensembles and aim to contribute to this area. Mainly, we provide two novel algorithms and corresponding lower bounds. First, we are able to carry over and substantially improve on tractability results for decision trees: We obtain an algorithm that, given a training-data set and an size bound $S \in \mathbb{R}$, computes a tree ensemble of size at most $S$ that classifies the data correctly. The algorithm runs in $(4δD S)^S \cdot poly$-time, where $D$ the largest domain size, $δ$ is the largest number of features in which two examples differ, $n$ the number of input examples, and $poly$ a polynomial of the input size. For decision trees, that is, ensembles of size 1, we obtain a running time of $(δD s)^s \cdot poly$, where $s$ is the size of the tree. To obtain these algorithms, we introduce the witness-tree technique, which seems promising for practical implementations. Secondly, we show that dynamic programming, which has been applied successfully to computing decision trees, may also be viable for tree ensembles, providing an $\ell^n \cdot poly$-time algorithm, where $\ell$ is the number of trees. Finally, we compare the number of cuts necessary to classify training data sets for decision trees and tree ensembles, showing that ensembles may need exponentially fewer cuts for increasing number of trees.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑