One of the most important functionalities in the ASTRID-system is the capability to use analogies while reasoning with uncertainty. The underlying ASTRID brain-model supports analogies, but obviously the inference engine needs to be able to use them. Today, the first implementation of this capability in the inference engine, showed its promise during early testing.

Analogies are a hard problem in Artificial Intelligence because it is a true 'chicken or the egg' problem. It has been argued (Chalmers, French, Hofstadter, 1991) that you need perception to be able to recognize analogies, and at the same time, analogies are fundamental for perception. For that reason, the system must be capable of building an analogy mapping, while it is using the analogy that is being built for perception of the incoming information.

This is specifically where the ASTRID-system excels. It's capability to find novel relational mappings between concepts, gives it the ability to build analogies from concepts while it is inferencing those concepts to aid in perception. Today we observed how the inference engine, being given two new but essentially analogous concepts, immediately recognized the similarities while actually learning these concepts for the first time.

Handling of analogous concepts is a primary necessity for fully autonomous systems, as it is needed to handle 'never seen before' situations. That's why this is such an important milestone for the ASTRID-project, as it brings us one step closer to an autonomously reasoning system.

NOTE: This news-article is presented here for historical perspective only.

This article is more than two years old. Therefore, information in this article might have changed, become incomplete, or even completely invalid since its publication date. Included weblinks (if present in this article) might point to pages that no longer exist, have been moved over time, or now contain unrelated or insufficient information. No expectations or conclusions should be derived from this article or any forward-looking statements therein.