Machine Learning for First-Order Theorem Proving Learning to Select a Good Heuristic

被引:52
|
作者
Bridge, James P. [1 ]
Holden, Sean B. [1 ]
Paulson, Lawrence C. [1 ]
机构
[1] Univ Cambridge, Comp Lab, Cambridge CB3 0FD, England
基金
英国工程与自然科学研究理事会;
关键词
Automatic theorem proving; Machine learning; First-order logic with equality; Feature selection; Support vector machines; Gaussian processes; CLASSIFICATION; SAT;
D O I
10.1007/s10817-014-9301-5
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We applied two state-of-the-art machine learning techniques to the problem of selecting a good heuristic in a first-order theorem prover. Our aim was to demonstrate that sufficient information is available from simple feature measurements of a conjecture and axioms to determine a good choice of heuristic, and that the choice process can be automatically learned. Selecting from a set of 5 heuristics, the learned results are better than any single heuristic. The same results are also comparable to the prover's own heuristic selection method, which has access to 82 heuristics including the 5 used by our method, and which required additional human expertise to guide its design. One version of our system is able to decline proof attempts. This achieves a significant reduction in total time required, while at the same time causing only a moderate reduction in the number of theorems proved. To our knowledge no earlier system has had this capability.
引用
收藏
页码:141 / 172
页数:32
相关论文
共 50 条
  • [41] Learning first-order probabilistic models with combining rules
    Sriraam Natarajan
    Prasad Tadepalli
    Thomas G. Dietterich
    Alan Fern
    Annals of Mathematics and Artificial Intelligence, 2008, 54 : 223 - 256
  • [42] First-order concept learning using genetic algorithms
    Tamaddoni-Nezhad, A
    Muggleton, S
    IC-AI'2001: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOLS I-III, 2001, : 688 - 694
  • [43] First-Order Optimization (Training) Algorithms in Deep Learning
    Rudenko, Oleg
    Bezsonov, Oleksandr
    Oliinyk, Kyrylo
    COMPUTATIONAL LINGUISTICS AND INTELLIGENT SYSTEMS (COLINS 2020), VOL I: MAIN CONFERENCE, 2020, 2604
  • [44] Margin-based first-order rule learning
    Rueckert, Ulrich
    Kramer, Stefan
    MACHINE LEARNING, 2008, 70 (2-3) : 189 - 206
  • [45] Margin-based first-order rule learning
    Ulrich Rückert
    Stefan Kramer
    Machine Learning, 2008, 70 : 189 - 206
  • [46] Learning first-order rules: A rough set approach
    Stepaniuk, J
    Honko, P
    FUNDAMENTA INFORMATICAE, 2004, 61 (02) : 139 - 157
  • [47] Inducing spatial knowledge with a first-order learning system
    Nicoletti, MD
    Brennan, J
    INTELLIGENT SYSTEMS, 2000, : 131 - 134
  • [48] Margin-based first-order rule learning
    Rueckert, Ulrich
    Kramer, Stefan
    INDUCTIVE LOGIC PROGRAMMING, 2007, 4455 : 46 - +
  • [49] First-order Logic Learning in Artificial Neural Networks
    Guillame-Bert, Mathieu
    Broda, Krysia
    Garcez, Artur d'Avila
    2010 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS IJCNN 2010, 2010,
  • [50] Learning first-order probabilistic models with combining rules
    Natarajan, Sriraam
    Tadepalli, Prasad
    Dietterich, Thomas G.
    Fern, Alan
    ANNALS OF MATHEMATICS AND ARTIFICIAL INTELLIGENCE, 2008, 54 (1-3) : 223 - 256