Rob, a lot of your disagreements stem from your language-first mindset. Which is perverse, you must agree that the language is a product of basic cognitive ability, possed by all mammals. Starting from your "contradiction": that's simply a linguistic equivalent of my variance. I have no idea why you think my nested hierarchies don't "appear naturally, in real time", and are not context-dependent? Maybe it's because you never tried to actually implement your ideas, so you don't think procedurally? I never said that all parerns must be global, the whole idea of hierarchy means that there is gradial transition from local to global.
>Compared to what I remember you are addressing "lateral" patterns now. Perhaps as a consequence of the success of transformers? It's always been lateral, I didn't learn anything worth borrowing from transformers, or neural nets in general. The whole idea of statistical fitting that underlies it is backwards, we do it because it's simple and we are lazy. Incremental functional complexity is what progress is all about. Evolution found some quick and dirty fixes: neural nets of whatever kind, now we need to use them to create something way better. Self-improvement. ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T682a307a763c1ced-Md375cd0b433d4b19f4dcb0e5 Delivery options: https://agi.topicbox.com/groups/agi/subscription