2021/10/07 by Laura Giordano, Giordano, Laura
Computer Science · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #I.2.4 #Logic, Reasoning, and Knowledge #Multi-Agent Systems and Negotiation #Natural Language Processing Techniques #Topic Modeling
paper · pdf · doi:10.48550/arxiv.2110.03643
openalex publication_date 2021/10/07 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
A fuzzy multipreference semantics has been recently proposed for weighted\nconditional knowledge bases, and used to develop a logical semantics for\nMultilayer Perceptrons, by regarding a deep neural network (after training) as\na weighted conditional knowledge base. This semantics, in its different\nvariants, suggests some gradual argumentation semantics, which are related to\nthe family of the gradual semantics studied by Amgoud and Doder. The\nrelationships between weighted conditional knowledge bases and MLPs extend to\nthe proposed gradual semantics to capture the stationary states of MPs, in\nagreement with previous results on the relationship between argumentation\nframeworks and neural networks. The paper also suggests a simple way to extend\nthe proposed semantics to deal attacks/supports by a boolean combination of\narguments, based on the fuzzy semantics of weighted conditionals, as well as an\napproach for defeasible reasoning over a weighted argumentation graph, building\non the proposed gradual semantics.\n