Starts at 04:10

It seems like there are 3 main paradigms when it comes to Generalized A.I.: 1) Those systems that are built piece-meal/in-tandem with discoveries about how actual neurons function/their theoretical operation 2) Statistical-Mechanical (forgive my loose vocabulary) systems that utilize Bayes Theorem (just as an example) along with large data models. (Though these tend to be treated like "Black boxes", so mysterious even the GAI software authors/engineers don't know exactly how the GAI internal "thoughts" should be interpreted/how they're being utilized 3) Abstract/Epistomo-logico Models. I've just started reading David Hume's "An Inquiry Concerning Human Understanding". I'm only a few sections in, and this might be a premature assessment, but it seems that the "mechanisms"(not Hume's vocabulary btw) of human thought Hume presents in the book could serve as a very interesting lattice/functional-framework for operations of individual/groups of neural-nets to provide ¿emergent? human-like intelligence. I'm not aware of.anyone who.has tried.this. I understand that Hume's framework doesn't provide for the "engine" so-to-speak of desire/goal-seeking-behavior you need in autonomous software brains. What im getting at is Hume's (which he argues is 'exhaustive') enumeration of the various ways we infer, say, cause&effect LOOKs like it could provide a fairly solid theoretical undermining for GAI thinking. Anybody out there know what I'm talking about?

Thumbnail is awesome something that knows all about this stuff would look a lil yah know

Why not making the video title more reachable and findable . It would be much better if you make the title such that it reflects the topic of the presentation

The velocity vector and the position vector will always be 2-dimensional or their dimensions depend on the amount of data to be modeled

The audio is a little bit off...

Great talk!