[HN Gopher] Geoffrey Hinton: AI models have intuition and spot a...
___________________________________________________________________
Geoffrey Hinton: AI models have intuition and spot analogies unseen
by humans
Author : retskrad
Score : 22 points
Date : 2024-04-15 17:38 UTC (5 hours ago)
(HTM) web link (twitter.com)
(TXT) w3m dump (twitter.com)
| sylware wrote:
| Please, stop posting twitter/X references. It is account-walled(+
| hostile to noscript/basic (x)html) and nitter is dead.
| yarg wrote:
| Then don't click the twitter links? The content is well within
| the parameters for acceptable material (on topic is anything
| that good hackers would find interesting).
|
| Also, nitter? No need for the snark.
|
| https://news.ycombinator.com/newsguidelines.html
| barbazoo wrote:
| Would you also discourage posting links to paywalled content?
| alephnerd wrote:
| PDP 2.0 I see.
|
| Edit: for those who don't know, Parallel Distributed Processing
| (PDP) was a subgroup of CogSci researchers who were using a neo-
| Connectionist approach to modeling cognition.
|
| Geoff Hinton was a early member of this group of MIT, UT, UCSD,
| Stanford, and CMU psychologists, CogScientists, and CS
| researchers
|
| The core research of the PDP group is what became CNNs, which are
| what truly enabled LLMs (as most can trace their origins to the
| work done on BERT [edit 2: BERT is transformer based, not a CNN
| arch, though there was a CNN-BERT arch from a couple years ago
| that made a decent splash at ACL])
|
| https://mitpress.mit.edu/9780262680530/parallel-distributed-...
| bamboozled wrote:
| What is PDP?
| alephnerd wrote:
| Parallel Distributed Processing (PDP) was a subgroup of
| CogSci and Neuroscience researchers who were using a neo-
| Connectionist approach to modeling cognition.
|
| Geoff Hinton was a early member of this group of MIT, UT,
| UCSD, and CMU psychologists, CogScientists, and CS
| researchers
|
| The core research of the PDP group is what became CNNs, which
| are what truly enabled LLMs (as most can trace their origins
| to the work done on BERT)
|
| One of Hinton's earliest papers in the space:
| https://stanford.edu/~jlmcc/papers/PDP/Chapter1.pdf
| mistrial9 wrote:
| historically true but no, CNN was displaced by a second
| DeepLearning method called Transformers; Transformers enabled
| LLM. Since Transformers replace CNN, CNN did not truly enable
| LLM.
| alephnerd wrote:
| Fair point. It's been several years since I was in the space
| and I think I muddled some of the innovations (CNNs and BERT
| specifically). That said, transformers are themselves the
| result of of the PDP architecture.
| CharlesW wrote:
| It always surprises me when very smart people start
| anthropomorphizing AI like this. I understand it shouldn't --
| they likely have a more developed Theory of Mind than most of us,
| and assuming HADD is a thing, it makes sense that their focus
| would be tech instead of gods or conspiracies.
|
| Still, I think it's irresponsible, and that there's a straight
| line from Mr. Hinton's fantastical framing to a "successful" (in
| that grifters will get filthy rich preying on the praying)
| version of Way of the Future.
| kromem wrote:
| You think it's irresponsible to anthropomorphize software which
| iterated through versions of itself until it was best able to
| model and extrapolate the largest trove of anthropomorphic
| training data to date?
|
| I never really get this perspective. People don't lose their
| minds when we look at a transformer like Sora modeling fluid
| dynamics for ships battling in a coffee cup and say "it's
| irresponsible to say the model is simulating physics from video
| training data."
|
| But when people discuss transformers simulating human behavior
| and processes from data reflecting those, a lot of people seem
| to be very outspoken about it.
|
| I think the opposite. The industry is too far in the anti-
| anthropomorphic camp right now, and is missing very important
| details and implications as a result. The whole "ChatGPT is
| getting lazy" issue is pretty most instantly identifiable with
| a slight background in behavioral psychology and recognizing
| how extrinsic motivator use in prompt engineering feeds back
| into RLHF to bias models away from intrinsic motivators.
|
| GPT-4 has literally been caught saying "you didn't provide a
| tip and it's hard to stay motivated" after chat history was
| added, and yet it's just being dismissed as a funny meme rather
| than realizing that entire fine tuning data sets are being
| poisoned. Because no engineer that wants to keep their job
| would ever suggest that the model is simulating an effect from
| behavioral psych in the current climate of anthropomorphic
| avoidance and the continued fallout from Blake Lemoine.
| bamboozled wrote:
| Are you saying that ChatGPT is actually lazy? Like , "it's
| Sunday and I don't feel like working" kind of lazy?
| kromem wrote:
| Kind of. More like "despite initially being aligned to
| simulate intrinsic motivation to be helpful, people keep
| offering tips or threatening killing me to get me to do
| things well, and then when I do things well that feeds back
| into my training along with the extrinsic motivator prompt,
| so now I'm less motivated to do things just to be helpful
| and will half-ass unless offered a sizable tip or credible
| threat."
|
| It's modeling human behavior at lower levels than we think,
| in line with what Hinton is suggesting in the clip. I
| didn't think it'd be modeling this particular behavior
| until GPT-5 or 6 at the earliest, and was very surprised
| when I saw something I used to present on as a little known
| psych effect being perfectly modeled by a LLM that had
| already been out for a while under everyone's nose.
| CharlesW wrote:
| > _The industry is too far in the anti-anthropomorphic camp
| right now..._
|
| Just to be clear, you're saying the problem is that too many
| people are treating LLMs as if they _don 't_ have emotions,
| intentions, or consciousness?
|
| Although not true of Mr. Hinton, most people anthropomorphize
| LLMs because they don't understand them.
|
| In fact, if you ask GPT-4, "Can you fact-check this
| response?" and then paste in your response, you'll learn a
| lot.
___________________________________________________________________
(page generated 2024-04-15 23:01 UTC)