arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1602.08610cs.AI
收藏

可扩展的贝叶斯规则列表

Scalable Bayesian Rule Lists

Massachusetts Institute of Technology(麻省理工学院);Duke University(杜克大学);Harvard University(哈佛大学)

Hongyu Yang, Cynthia Rudin, Margo Seltzer

摘要:

我们提出了一种算法,用于建立概率规则列表,这是两个数量级的速度比以前的工作。规则列表算法是决策树算法的竞争对手。它们是关联分类器,因为它们是从预先挖掘的关联规则中构建的。它们的逻辑结构是一系列“若-则”规则,与决策列表或单边决策树相同。而不是使用贪婪的分裂和修剪一样的决策树算法,我们充分优化规则列表,在准确性,可解释性和计算速度之间取得了实际的平衡。这里提出的算法使用理论边界的混合(足够紧密,具有实际意义的筛选或边界过程),计算重用,和高度调优的语言库,以实现计算效率。目前,对于许多实际问题,该方法比决策树具有更好的准确性和稀疏性;此外,在许多情况下,计算时间是实用的,并且通常小于决策树。结果是概率分类器(其估计P(y = 1|对于每个x),优化贝叶斯分层模型在规则列表上的后验。

英文摘要:

We present an algorithm for building probabilistic rule lists that is two orders of magnitude faster than previous work. Rule list algorithms are competitors for decision tree algorithms. They are associative classifiers, in that they are built from pre-mined association rules. They have a logical structure that is a sequence of IF-THEN rules, identical to a decision list or one-sided decision tree. Instead of using greedy splitting and pruning like decision tree algorithms, we fully optimize over rule lists, striking a practical balance between accuracy, interpretability, and computational speed. The algorithm presented here uses a mixture of theoretical bounds (tight enough to have practical implications as a screening or bounding procedure), computational reuse, and highly tuned language libraries to achieve computational efficiency. Currently, for many practical problems, this method achieves better accuracy and sparsity than decision trees; further, in many cases, the computational time is practical and often less than that of decision trees. The result is a probabilistic classifier (which estimates P(y = 1|x) for each x) that optimizes the posterior of a Bayesian hierarchical model over rule lists.

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。

↑