A Semantic Loss Function for Deep Learning with Symbolic Knowledge
A Semantic Loss Function for Deep Learning with Symbolic Knowledge
复制标题
DOI：
--
复制
发表时间：
2017-11
期刊：
影响因子：
--
通讯作者：
Jingyi Xu;Zilu Zhang;Tal Friedman;Yitao Liang;Guy Van den Broeck
中科院分区：
文献类型：
--
作者：
Jingyi Xu;Zilu Zhang;Tal Friedman;Yitao Liang;Guy Van den Broeck
本文提出了一种在深度学习中使用符号知识的新方法。从第一性原理出发，我们推导出连接神经输出向量和逻辑约束的语义损失函数。该损失函数捕捉到神经网络在多大程度上满足其输出约束。实验评估表明，该方法有效地指导学习者在半监督多类分类上获得(接近)最先进的结果。此外，它还显著提高了神经网络预测结构化对象(如排名和路径)的能力。这些离散的概念非常难以学习，并受益于深度学习和符号推理方法的紧密结合。
This paper develops a novel methodology for using symbolic knowledge in deep learning. From first principles, we derive a semantic loss function that bridges between neural output vectors and logical constraints. This loss function captures how close the neural network is to satisfying the constraints on its output. An experimental evaluation shows that it effectively guides the learner to achieve (near-)state-of-the-art results on semi-supervised multi-class classification. Moreover, it significantly increases the ability of the neural network to predict structured objects, such as rankings and paths. These discrete concepts are tremendously difficult to learn, and benefit from a tight integration of deep learning and symbolic reasoning methods.