On Thu, May 23, 2024 at 10:10 AM Quan Tesla <quantes...@gmail.com> wrote: > > The paper is specific to a novel and quantitative approach and method for > association in general and specifically.
John was talking about the presentation James linked, not the paper, Quan. He may be right that in that presentation they use morphisms etc to map learned knowledge from one domain to another. He's not criticising the paper though. Only the presentation. And the two were discussing different techniques. John isn't criticising the Granger et al. "relational encoding" paper at all. > The persistence that pattern should be somehow decoupled doesn't make much > sense to me. Information itself is as a result of pattern. Pattern is > everything. Light itself is a pattern, so are the four forces. Ergo. I > suppose, it depends on how you view it. If you're questioning my point, it is that definition in terms of relations means the pattern can vary. It's like the gap filler example in the paper: "If John kissed Mary, Bill kissed Mary, and Hal kissed Mary, etc., then a novel category ¢X can be abduced such that ¢X kissed Mary. Importantly, the new entity ¢X is not a category based on the features of the members of the category, let alone the similarity of such features. I.e., it is not a statistical cluster in any usual sense. Rather, it is a “position-based category,” signifying entities that stand in a fixed relation with other entities. John, Bill, Hal may not resemble each other in any way, other than being entities that all kissed Mary. Position based categories (PBCs) thus fundamentally differ from “isa” categories, which can be similarity-based (in unsupervised systems) or outcome-based (in supervised systems)." If you define your category on the basis of kissing Mary, then who's to say that you might not find other people who have kissed Mary, and change your category from moment to moment. As you discovered clusters of former lovers by fits and starts, the actual pattern of your "category" might change dramatically. But it would still be defined by its defining relation of having kissed Mary. That might also talk to the "regression" distinction. Or characterizing the system, or indeed all cognition, as "learning" period. It elides both "similarity-based" unsupervised, and supervised, "learning". The category can in fact grow as you "learn" of new lovers. A process which I also have difficulty equating with regression. ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T682a307a763c1ced-M8c58bf8eb0a279da79ea34eb Delivery options: https://agi.topicbox.com/groups/agi/subscription