On computable abstractions (a conceptual introduction)
This addresses the challenge of autonomous machine understanding for AI systems, but it appears conceptual and incremental in its approach.
The paper tackles the problem of enabling computers to autonomously build and use meaningful abstractions, introducing 'abstractional machines' that mechanically create their own understanding for intellectual tasks, with an example application to natural language processing.
This paper introduces abstractions that are meaningful for computers and that can be built and used according to computers' own criteria, i.e., computable abstractions. It is analyzed how abstractions can be seen to serve as the building blocks for the creation of one own's understanding of things, which is essential in performing intellectual tasks. Thus, abstractional machines are defined, which following a mechanical process can, based on computable abstractions, build and use their own understanding of things. Abstractional machines are illustrated through an example that outlines their application to the task of natural language processing.