5 ms·
> Some perceptual tasks work better than a human, in fact. [...] that's not something you can say about the last AI winter Eh. I'd say that's somewhat apples t
by ethbro 7y ago
> Some perceptual tasks work better than a human, in fact. [...] that's not something you can say about the last AI winter
Eh. I'd say that's somewhat apples to oranges.
A) There were some useful and successful expert systems.
B) Things seemed to be going swimmingly, until they hit a fundamental wall.
C) We're working with a few orders of magnitude greater compute than they had access to.
- m0zg 7y agoSure, but we did figure out how to make things more robust and generalizable, at least for perceptual tasks so far. Knowledge representation and probabilistic reasoning are still non-existent, though. Moreover, nobody is even working on any of that, for fear of being compared to Doug Lenat.
- snaky 7y agoSo robust and generalizable that adding a pixel here and a pixel there, too small to be even recognized by a human, confuse image recognition system to the point it confuse panda with gibbon?
- Quetelet 7y agoThese adversarial examples are generated in a very artificial way that will not be present in natural images (and if you’re thinking of security issues, the attacker needs access to your model...) They’re still an interesting topic to explore but hardly evidence that neural nets don’t generalize.
- goatlover 7y agoWouldn't the worry be that generalizing to real world tasks involves enough variation that the same issue could arise naturally?
- Quetelet 7y agoRepresentation learning and probabilistic methods are huge sub-areas of modern machine learning, just take a look at the proceedings of ICLR2019.
- m0zg 7y agoRepresentation learning != knowledge representation, probabilistic methods != probabilistic reasoning. I'm talking foundations of AGI, which as far as I'm aware, nobody is seriously working on at the moment.