An ASP Approach for Reasoning on Neural Networks under a Finitely Many-Valued Semantics for Weighted Conditional Knowledge Bases

LAURA GIORDANO; DANIELE THESEIDER DUPRE

首页> 外文期刊>Theory and practice of logic programming >An ASP Approach for Reasoning on Neural Networks under a Finitely Many-Valued Semantics for Weighted Conditional Knowledge Bases

【24h】

An ASP Approach for Reasoning on Neural Networks under a Finitely Many-Valued Semantics for Weighted Conditional Knowledge Bases

机译：An ASP Approach for Reasoning on Neural Networks under a Finitely Many-Valued Semantics for Weighted Conditional Knowledge Bases

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相关主题

摘要

Weighted knowledge bases for description logics with typicality have been recently consideredunder a “concept-wise” multipreference semantics (in both the two-valued and fuzzy case), as thebasis of a logical semantics of multilayer perceptrons (MLPs). In this paper we consider weightedconditional ALC knowledge bases with typicality in the finitely many-valued case, through threedifferent semantic constructions. For the boolean fragment LC of ALC we exploit answer setprogramming and asprin for reasoning with the concept-wise multipreference entailment undera ?-coherent semantics, suitable to characterize the stationary states of MLPs. As a proof ofconcept, we experiment the proposed approach for checking properties of trained MLPs.

著录项

来源
《Theory and practice of logic programming》 |2022年第4期|589-605|共17页
作者
LAURA GIORDANO; DANIELE THESEIDER DUPRE;
展开▼
作者单位

DISIT, Universita del Piemonte Orientale, Italy;

展开▼
收录信息
原文格式 PDF
正文语种英语
中图分类
关键词
description logics; neural networks; multi-valued logics; answer set programming;

An ASP Approach for Reasoning on Neural Networks under a Finitely Many-Valued Semantics for Weighted Conditional Knowledge Bases

摘要

著录项

相关主题

期刊订阅