Weighted defeasible knowledge bases and a multipreference semantics for a deep neural network model

TL;DR

This paper explores the connection between multipreferential semantics for defeasible reasoning in knowledge bases and deep neural network models, extending weighted description logics to fuzzy interpretations and applying them to multilayer perceptrons.

Contribution

It introduces a novel multipreference semantics framework for defeasible reasoning and extends it to neural network models, bridging knowledge representation and deep learning.

Findings

Extended weighted knowledge bases to fuzzy interpretations

Provided a preferential interpretation for multilayer perceptrons

Linked defeasible reasoning semantics with neural network models

Abstract

In this paper we investigate the relationships between a multipreferential semantics for defeasible reasoning in knowledge representation and a deep neural network model. Weighted knowledge bases for description logics are considered under a "concept-wise" multipreference semantics. The semantics is further extended to fuzzy interpretations and exploited to provide a preferential interpretation of Multilayer Perceptrons.