Jeroen Cottaar: prediction

Transcript context

…So like if you were an agent that genuinely needed to acquire knowledge to solve the task, so you had to do like the minimum number of steps to acquire the knowledge. So do you think that would converge on a certain score? Yeah. So that that's what I think would be like 90% or something. Yeah. Simply because it is not reasonable to to Like we just saw the RS 20 game, there's 2 places you can move to. Yeah. You you just can't guess that right every time. The thing is, you're being compared to humans who also had to do the same exploration and you get the median score. So kind of with a If you're averagely lucky, you're going to do it. But, yeah, you have a 110 games. Are you going to get unlucky in 1 of them? Yeah. And not yet I will mention, even with our simple With GWEN, we can already get sometimes 100% on certain games. So that it is, let's say, certainly doable What to beat those human we were saying earlier is that, 1 view is that intelligence is this quite crystallized process. There's a concept of IQ, for example, and if you have a certain amount of IQ, then it's predictable how well you can generalize, how efficient you are. And guess the data might just not back that up. The data might just say actually it's kind of not random but very specialised and there's huge differences, individual differences in capability and machines and in humans. And would that make them kind of reassess their whole idea of what intelligence is? On ArcGI 3 specifically,…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.