Re:Language models can explain neurons in language models
by necroforest on Tuesday May 09, 2023 @10:06PM
(#35882790)
> AI research has put hardly any effort into building goal-directed agents
The entire (enormous) field of reinforcement learning begs to differ.
Re:Language models can explain neurons in language models
by robwwilliams on Wednesday May 10, 2023 @12:44AM
(#35883728)
Brilliant comment—-and back to basics.
Yes, and put that compact fruit fly in silico brain into my Roomba please so that it does not get stuck under the bed.
This is the kind of embodied AI that should really worry us.
Don’t we all suspect deep skunkworks “defense” projects of these types?
Re:Language models can explain neurons in language models
by pmontra on Wednesday May 10, 2023 @05:54AM
(#35885471)
Well, flies and all sort of flying bugs are very good at getting into homes and very bad at finding a way out. They stick on a closed window and can't find the open one next to it.
Re:Language models can explain neurons in language models
by quincey on Wednesday May 10, 2023 @12:52PM
(#35890001)
There's no genetic advantage to "finding a way out"! The home barrier way in is a genetic hurdle - flies that cross it are free to reproduce in an abundant environment. This calls for a "quieter" fly (a stealth fly?) who annoys the local beasts minimally - yet another genetic hurdle.
Re:Language models can explain neurons in language models
by stormfather on Wednesday May 10, 2023 @11:11AM
(#35888518)
I wouldn't assume a single biological neuron is simple like an artificial neuron is.
Computationally, a biological neuron is equivalent to something like an entire artificial neural net.
I don't think there's a single dollar of funding in AI right now going to the "problem" of making an AI that has terminal preferences, sets instrumental goals to achieve those preferences, iteratively observes the environment and makes predictions of how well its actions will help optimize the future toward its preferences, etc.
Though, interestingly, it might not even be that hard a problem, compared to all the problems being solved in AI right now. A fruit fly is already a goal-directed agent in the sense described above. Yet a fruit fly has only 200K neurons, and very few of the connections between those neurons are dynamic; most are "hard wired" by [probably] genetics.
If we want true ALife, we only need to understand what a fruit fly brain is doing, and then model it. And that model will then fit — with room to spare! — on a single GPU. From a decade ago.