Learning Logic Programs Using Neural Networks by Exploiting Symbolic Invariance

Learning Logic Programs Using Neural Networks by Exploiting Symbolic Invariance
复制标题

利用符号不变性使用神经网络学习逻辑程序

DOI:
10.1007/978-3-030-97454-1_15
复制
发表时间:
2022
期刊:
In: Nikos Katzouris, Alexander Artikis (Eds.): Inductive Logic Programming, Proceedings of the 30th International Conference (ILP 2020-2021; Virtual Event), Lecture Notes in Artificial Intelligence
影响因子:
--
通讯作者:
Katsumi Inoue
Katsumi Inoue
中科院分区:
--
文献类型:
--
作者:
Yin Jun Phua;Katsumi Inoue

文献摘要

参考文献

相似文献

从解释转换中学习(LFIT)是一种无监督学习算法,它只通过观察状态转换来学习动态。LFIT算法主要以符号方法实现,但它们对噪声或丢失数据不鲁棒。最近,将逻辑运算与神经网络相结合的研究工作受到了很多关注,其中大多数工作采用基于提取的方法,其中训练单个神经网络模型来解决问题,然后从神经网络模型中提取逻辑模型。然而,大多数研究工作遭受组合爆炸问题时,试图扩大规模,以解决更大的问题。特别是符号世界中的许多不变性在神经网络领域没有得到利用。在这项工作中,我们提出了一个模型,利用符号不变性在我们的问题。我们表明,我们的模型能够扩展到更大的任务比以前的工作。
Learning from Interpretation Transition (LFIT) is an unsupervised learning algorithm which learns the dynamics just by observing state transitions. LFIT algorithms have mainly been implemented in the symbolic method, but they are not robust to noisy or missing data. Recently, research works combining logical operations with neural networks are receiving a lot of attention, with most works taking an extraction based approach where a single neural network model is trained to solve the problem, followed by extracting a logic model from the neural network model. However most research work suffer from the combinatorial explosion problem when trying to scale up to solve larger problems. In particular a lot of the invariance that hold in the symbolic world are not getting utilized in the neural network field. In this work, we present a model that exploits symbolic invariance in our problem. We show that our model is able to scale up to larger tasks than previous work.
DOI: 10.3389/fbioe.2014.00081
发表时间: 2014
影响因子: 5.7
作者:
Ribeiro T;Magnin M;Inoue K;Sakama C
通讯作者: Sakama C
开发 Adaptive Boost 方法并使用该方法分析铁中的碳扩散 [MD 奖获得者]
DOI: --
发表时间: 2011
期刊:
影响因子: --
作者:
石井明男;尾方成信;君塚肇
通讯作者: 君塚肇
使用经过训练的神经网络的布尔网络提取,从时间序列数据中对基因调控网络进行逆向工程
DOI: --
发表时间: 2019
期刊:
影响因子: --
作者:
T. Rintala
通讯作者: T. Rintala