Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Learning from context is harder than we thought

hy.tencent.com

Post reply on HN
Print view View original

81–90 of 140 posts

Re: Learning from context is harder than we thought

#81

Post by pankajdoharey »

bluegatty wrote: post #61
observationist wrote: post #39

Earlier quoted context omitted.

It 100% needs to be online. Imagine you're trying to think about a new tabletop puzzle, and every time a puzzle piece leaves your direct field of view, you no longer know about that puzzle piece. You can try to keep all of the puzzle pieces within your direct field of view, but that divides your focus. You can hack that and make your field of view incredibly large, but that can potentially distort your sense of the r…

That's not how training works - adjusting model weights to memorize a single data item is not going to fly. Model weights store abilities, not facts - generally. Unless the fact is very widely used and widely known, with a ton of context around it. The model can learn the day JFK died because there are millions of sparse examples of how that information exists in the world, but when you're working on a problem, you m…

Not to forget we will need thousands of examples for the models to extract abilities the sample efficiency of these models is quite poor.

Re: Learning from context is harder than we thought

#83

Post by red75prime »

numtel wrote: post #16
Tor3 wrote: post #14

Earlier quoted context omitted.

"There are examples of tribal humans not being able to perceive a green square among blue squares, because their language does not have a word for the green color. Similarly, some use the same word for blue and white, and are unable to perceive them as different colors." Both of the above is false. There are a ton of different colors that I happen to call "red", that does not mean that I can't perceive them as differ…

There's a Terence McKenna quote about this: > So, for instance, you know, I’ve made this example before: a child lying in a crib and a hummingbird comes into the room and the child is ecstatic because this shimmering iridescence of movement and sound and attention, it’s just wonderful. I mean, it is an instantaneous miracle when placed against the background of the dull wallpaper of the nursery and so forth. But, the…

Haha. I'd prefer for him to dance this sentence or something. To not detract from the marvel of being with crude words.

Re: Learning from context is harder than we thought

#84

Post by Nevermark »

com2kid wrote: post #35
4b11b4 wrote: post #31

Earlier quoted context omitted.

I'm not sure if you want models perpetually updating weights. You might run into undesirable scenarios.

If done right, one step closer to actual AGI. That is the end goal after all, but all the potential VCs seem to forget that almost every conceivable outcome of real AGI involves the current economic system falling to pieces. Which is sorta weird. It is like if VCs in Old Regime france started funding the revolution.

If it makes the models smarter, someone will do it.

From any individual, up to entire countries, not participating doesn't do anything except ensure you don't have a card to play when it happens.

There is a very strong element of the principles of nature and life (as in survival, not nightclubs or hobbies) happening here that can't be shamed away.

The resource feedback for AI progress effort is immense (and it doesn't matter how much is earned today vs. forward looking investment). Very few things ever have that level of relentless force behind them. And even beyond the business need, keeping up is rapidly becoming a security issue for everyone.

Re: Learning from context is harder than we thought

#85

Post by DoctorOetker »

pankajdoharey wrote: post #80
embedding-shape wrote: post #38

Earlier quoted context omitted.

> We need models that keep on learning (updating their parameters) forever, online, all the time. Do we need that? Today's models are already capable in lots of areas. Sure, they don't match up to what the uberhypers are talking up, but technology seldom does. Doesn't mean what's there already cannot be used in a better way, if they could stop jamming it into everything everywhere.

Continuous learningin current models will lead to catastrophic forgetting.

will catastrophic forgetting still occur if a fraction of the update sentences are the original training corpus?

is the real issue actually catastrophic forgetting or overfitting?

nothing prevents users from continuing the learning as they use a model

Re: Learning from context is harder than we thought

#86

Post by snayan »

numtel wrote: post #16
Tor3 wrote: post #14

Earlier quoted context omitted.

"There are examples of tribal humans not being able to perceive a green square among blue squares, because their language does not have a word for the green color. Similarly, some use the same word for blue and white, and are unable to perceive them as different colors." Both of the above is false. There are a ton of different colors that I happen to call "red", that does not mean that I can't perceive them as differ…

There's a Terence McKenna quote about this: > So, for instance, you know, I’ve made this example before: a child lying in a crib and a hummingbird comes into the room and the child is ecstatic because this shimmering iridescence of movement and sound and attention, it’s just wonderful. I mean, it is an instantaneous miracle when placed against the background of the dull wallpaper of the nursery and so forth. But, the…

I think about this often. I've really come to appreciate over the past year the ways language can limit and warp our perception of reality. I think we under appreciate preverbal thought, as it seems to me that verbal thought by it's very nature has passed through our egoic filter, and our perception tends to be biased by our previous lived experience.

Socrates, Einstein, Nietzsche, Mozart.... So many of the greats described some of their most brilliant flashes of inspiration as just having come to them. Einstein's line about pure logical thinking not yielding knowledge of the emperical world, I really think these guys were good at daydreaming and able to tap into some part of themselves where intuition and preverbal thought could take the wheel, from which inspiration would strike.

Re: Learning from context is harder than we thought

#87

Post by logicchains »

johnsmith1840 wrote: post #3

It's basically continual learning. This is beyond a hard problem it's currently an impossible one. I know of no system that solve CL even at small scale let alone large models. Annoyingly, they have SOME inherent capability to do it. It's really easy to get sucked down this path due to that glimmer of hope but the longer you play with it the more annoying it becomes. SSI seems to be focused on this problem directly s…

Schmidhuber solved it at a small scale: https://arxiv.org/abs/2202.05780 .

Re: Learning from context is harder than we thought

#88

Post by fhd2 »

furyofantares wrote: post #67
cs702 wrote: post #24

The problem is even more fundamental: Today's models stop learning once they're deployed to production. There's pretraining, training, and finetuning, during which model parameters are updated. Then there's inference, during which the model is frozen. "In-context learning" doesn't update the model. We need models that keep on learning (updating their parameters) forever, online, all the time.

Why is learning an appropriate metaphor for changing weights but not for context? There are certainly major differences in what they are good or bad at and especially how much data you can feed them this way effectively. They both have plenty of properties we wish the other had. But they are both ways to take an artifact that behaves as if it doesn't know something and produce an artifact that behaves as if it does.…

I suppose ultimately, the external behaviour of the system is what matters. You can see the LLM as the system, on a low level, or even the entire organisation of e.g. OpenAI at a high level.

If it's the former: Yeah, I'd argue they don't "learn" much (!) past inference. I'd find it hard to argue context isn't learning at all. It's just pretty limited in how much can be learned post inference.

If you look at the entire organisation, there's clearly learning, even if relatively slow with humans in the loop. They test, they analyse usage data, and they retrain based on that. That's not a system that works without humans, but it's a system that I would argue genuinely learns. Can we build a version of that that "learns" faster and without any human input? Not sure, but doesn't seem entirely impossible.

Do either of these systems "learn like a human"? Dunno, probably not really. Artificial neural networks aren't all that much like our brains, they're just inspired by them. Does it really matter beyond philosophical discussions?

I don't find it too valuable to get obsessed with the terms. Borrowed terminology is always a bit off. Doesn't mean it's not meaningful in the right context.

Re: Learning from context is harder than we thought

#89

Post by robochat »

4b11b4 wrote: post #75
0xdeadbeefbabe wrote: post #44

Earlier quoted context omitted.

How about we just put them to bed once in a while?

Please elaborate on this one

I think they mean that the model should have sleep period where they update themselves with what they learnt that day.

Re: Learning from context is harder than we thought

#90

Post by nstart »

cs702 wrote: post #24

The problem is even more fundamental: Today's models stop learning once they're deployed to production. There's pretraining, training, and finetuning, during which model parameters are updated. Then there's inference, during which the model is frozen. "In-context learning" doesn't update the model. We need models that keep on learning (updating their parameters) forever, online, all the time.

Is this correct? My assumption is that all the data collected during usage is part of the RLHF loop of LLM providers. Assumption is based on information from books like empire of ai which specifically mention intent of AI providers to train/tune their models further based on usage feedback (eg: whenever I say the model is wrong in its response, thats a human feedback which gets fed back into improving the model).
Post reply on HN