JSAI2025

Presentation information

General Session

General Session » GS-1 Fundamental AI, theory

[2L4-GS-1] Fundamental AI, theory, algorithm:

Wed. May 28, 2025 1:40 PM - 3:20 PM Room L (Room 1007)

座長:中口 悠輝(NEC)

2:00 PM - 2:20 PM

[2L4-GS-1-02] Logical reasoning using graphical networks and its refinement using category theory

〇Yoshiki Fukada1 (1. Toyota Motor Corporation)

Keywords:category theory, logical reasoning, database, graph

Vast progresses are seen in neural-network based artificial intelligences. However, limitations are pointed out concerning their abilities of logical reasoning. This research presents a framework for logical reasoning using graphical networks: cognitive-logs. Cognitive-logs is a set of databases where the main elements are “actions” represented by verbs in natural languages and “participants” who perform the actions. The construction of cognitive-logs refers cognitive linguistics so that it represents humans cognitive mechanisms. By applying category theory, cognitive-logs flexibly reasons with logical rigors. This presentation illustrates the construction of cognitive-logs and demonstrates the wide applicability of cognitive-logs. Recognizing actions as entities was the key to the construction. Computations into cognitive-logs reproduce theories of cognitive sciences such as prototype theory.

Authentication for paper PDF access
A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.

Password