Deductive and Analogical Reasoning on a Semantically Embedded Knowledge Graph
This work addresses the challenge of robust and flexible reasoning for AI systems, though it appears incremental as it builds on existing vector space representations.
The paper tackles the problem of performing deductive reasoning in a continuous semantic vector space to address brittleness and incompleteness in knowledge bases, resulting in a method that integrates analogy, association, and deduction for reasoning chains using diverse knowledge sources.
Representing knowledge as high-dimensional vectors in a continuous semantic vector space can help overcome the brittleness and incompleteness of traditional knowledge bases. We present a method for performing deductive reasoning directly in such a vector space, combining analogy, association, and deduction in a straightforward way at each step in a chain of reasoning, drawing on knowledge from diverse sources and ontologies.