Ultra-Strong Machine Learning: comprehensibility of programs learned with ILP

Ultra-Strong Machine Learning: comprehensibility of programs learned with ILP
复制标题

超强机器学习:通过 ILP 学习的程序的可理解性

DOI:
10.1007/s10994-018-5707-3
复制
发表时间:
2018
期刊:
影响因子:
7.5
通讯作者:
Muggleton S
Muggleton S
中科院分区:
计算机科学3区
文献类型:
--
作者:
Muggleton S

文献摘要

参考文献

被引文献

相似文献

在20世纪80年代,Michie根据两个正交的性能轴定义了机器学习:预测准确性和生成假设的可理解性。由于预测准确性是容易测量的,而可理解性则不是这样,20世纪90年代的后期定义,如米切尔的定义,倾向于使用基于预测准确性的一维机器学习方法,最终倾向于统计而不是符号机器学习方法。在本文中,我们提供了假设的可理解性的定义,它可以通过人体参与试验来估计。我们提出了两组测试人类逻辑程序可理解性的实验。在第一个实验中,我们测试了有和没有谓词发明的人的可理解性。结果表明,可理解性不仅受到程序复杂性的影响,还受到匿名谓词符号存在的影响。在第二个实验中,我们直接测试了是否有任何最先进的ILP系统是Michie意义上的超强学习者,并选择Metagol系统用于人体试验。结果表明,参与者不能从一组例子中自己学习关系概念,但他们能够正确地应用ILP系统提供的关系定义。这意味着存在一类关系概念,这些概念对人类来说很难获得,尽管通过抽象的解释很容易理解。我们相信,提高对这门课的理解可能与人类学习、教学和语言互动的背景有潜在的关联。
During the 1980s Michie defined Machine Learning in terms of two orthogonal axes of performance: predictive accuracy and comprehensibility of generated hypotheses. Since predictive accuracy was readily measurable and comprehensibility not so, later definitions in the 1990s, such as Mitchell’s, tended to use a one-dimensional approach to Machine Learning based solely on predictive accuracy, ultimately favouring statistical over symbolic Machine Learning approaches. In this paper we provide a definition of comprehensibility of hypotheses which can be estimated using human participant trials. We present two sets of experiments testing human comprehensibility of logic programs. In the first experiment we test human comprehensibility with and without predicate invention. Results indicate comprehensibility is affected not only by the complexity of the presented program but also by the existence of anonymous predicate symbols. In the second experiment we directly test whether any state-of-the-art ILP systems are ultra-strong learners in Michie’s sense, and select the Metagol system for use in humans trials. Results show participants were not able to learn the relational concept on their own from a set of examples but they were able to apply the relational definition provided by the ILP system correctly. This implies the existence of a class of relational concepts which are hard to acquire for humans, though easy to understand given an abstract explanation. We believe improved understanding of this class could have potential relevance to contexts involving human learning, teaching and verbal interaction.
递归 Prolog 定义的有效归纳
DOI: --
发表时间: 1996
期刊: Canadian Conference on AI
影响因子: --
作者:
Riverson Rios;S. Matwin
通讯作者: S. Matwin
函数式程序的归纳综合:一种基于解释的泛化方法
DOI: 10.1023/a:1008797606116
发表时间: 2006
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
E. Kitzelmann;Ute Schmid
通讯作者: Ute Schmid
为什么和为什么不解释可以提高上下文感知智能系统的可理解性
DOI: 10.1145/1518701.1519023
发表时间: 2009
期刊: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
影响因子: --
作者:
Brian Y. Lim;A. Dey;Daniel Avrahami
通讯作者: Daniel Avrahami
通过饱和分析归纳具有多个递归子句的逻辑程序
DOI: --
发表时间: 1997
期刊: International Conference on Inductive Logic Programming
影响因子: --
作者:
Mitsue Furusawa;Nobuhiro Inuzuka;H. Seki;H. Itoh
通讯作者: H. Itoh
谓词发明如何影响人类的可理解性?
DOI: --
发表时间: 2016
期刊: International Conference on Inductive Logic Programming
影响因子: --
作者:
Ute Schmid;Christina Zeller;Tarek R. Besold;Alireza Tamaddoni;S. Muggleton
通讯作者: S. Muggleton