arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2006.03626cs.AI
收藏

LGML:逻辑引导的机器学习

LGML: Logic Guided Machine Learning

University of Waterloo(滑铁卢大学)

Joseph Scott, Maysum Panju, Vijay Ganesh

摘要:

我们介绍了逻辑引导机器学习(LGML),一种新的方法,共生结合机器学习(ML)和逻辑求解器的目标是从数据中学习数学函数。LGML由两个阶段组成,即学习阶段和具有校正反馈回路的逻辑阶段,使得学习阶段从输入数据学习符号表达式,并且逻辑阶段交叉验证学习的表达式与已知辅助真理的一致性。如果不一致,逻辑阶段将“反例”反馈给学习阶段。重复这个过程,直到学习的表达式与辅助真理一致。使用LGML,我们能够学习对应于勾股定理和正弦函数的表达式,与基于开箱即用的多层感知器(MLP)的方法相比,数据效率提高了几个数量级。

英文摘要:

We introduce Logic Guided Machine Learning (LGML), a novel approach that symbiotically combines machine learning (ML) and logic solvers with the goal of learning mathematical functions from data. LGML consists of two phases, namely a learning-phase and a logic-phase with a corrective feedback loop, such that, the learning-phase learns symbolic expressions from input data, and the logic-phase cross verifies the consistency of the learned expression with known auxiliary truths. If inconsistent, the logic-phase feeds back "counterexamples" to the learning-phase. This process is repeated until the learned expression is consistent with auxiliary truth. Using LGML, we were able to learn expressions that correspond to the Pythagorean theorem and the sine function, with several orders of magnitude improvements in data efficiency compared to an approach based on an out-of-the-box multi-layered perceptron (MLP).

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。

↑