Learning logic programs by explaining their failures

被引:0
|
作者
Morel, Rolf [1 ]
Cropper, Andrew [1 ]
机构
[1] Univ Oxford, Oxford, England
基金
英国工程与自然科学研究理事会;
关键词
Relational learning; Inductive logic programming; Failure explanation;
D O I
10.1007/s10994-023-06358-1
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Scientists form hypotheses and experimentally test them. If a hypothesis fails (is refuted), scientists try to explain the failure to eliminate other hypotheses. The more precise the failure analysis the more hypotheses can be eliminated. Thus inspired, we introduce failure explanation techniques for inductive logic programming. Given a hypothesis represented as a logic program, we test it on examples. If a hypothesis fails, we explain the failure in terms of failing sub-programs. In case a positive example fails, we identify failing sub-programs at the granularity of literals. We introduce a failure explanation algorithm based on analysing branches of SLD-trees. We integrate a meta-interpreter based implementation of this algorithm with the test-stage of the Popper ILP system. We show that fine-grained failure analysis allows for learning fine-grained constraints on the hypothesis space. Our experimental results show that explaining failures can drastically reduce hypothesis space exploration and learning times.
引用
收藏
页码:3917 / 3943
页数:27
相关论文
共 50 条
  • [1] Learning logic programs by explaining their failures
    Rolf Morel
    Andrew Cropper
    [J]. Machine Learning, 2023, 112 : 3917 - 3943
  • [2] Reporting Failures in Functional Logic Programs
    Hanus, Michael
    [J]. ELECTRONIC NOTES IN THEORETICAL COMPUTER SCIENCE, 2007, 177 (01) : 59 - 73
  • [3] Learning programs by learning from failures
    Andrew Cropper
    Rolf Morel
    [J]. Machine Learning, 2021, 110 : 801 - 856
  • [4] Learning programs by learning from failures
    Cropper, Andrew
    Morel, Rolf
    [J]. MACHINE LEARNING, 2021, 110 (04) : 801 - 856
  • [5] Misestimation: Explaining US Failures to Predict Nuclear Weapons Programs
    Montgomery, Alexander H.
    Mount, Adam
    [J]. INTELLIGENCE AND NATIONAL SECURITY, 2014, 29 (03) : 357 - 386
  • [6] Learning efficient logic programs
    Cropper, Andrew
    Muggleton, Stephen H.
    [J]. MACHINE LEARNING, 2019, 108 (07) : 1063 - 1083
  • [7] Learning efficient logic programs
    Andrew Cropper
    Stephen H. Muggleton
    [J]. Machine Learning, 2019, 108 : 1063 - 1083
  • [8] LEARNING RELATIONS AND LOGIC PROGRAMS
    BERGADANO, F
    GUNETTI, D
    [J]. KNOWLEDGE ENGINEERING REVIEW, 1994, 9 (01): : 73 - 77
  • [9] Learning extended logic programs
    Inoue, K
    Kudoh, Y
    [J]. IJCAI-97 - PROCEEDINGS OF THE FIFTEENTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOLS 1 AND 2, 1997, : 176 - 181
  • [10] Learning Efficient Logic Programs
    Cropper, Andrew
    [J]. PROCEEDINGS OF THE TWENTY-FOURTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI), 2015, : 4359 - 4360