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

DeepMind: A Generalist Agent

deepmind.com

Post reply on HN
Print view View original

291–300 of 345 posts

Re: DeepMind: A Generalist Agent

#291

Post by ben_w »

brador wrote: post #283
ben_w wrote: post #220

Earlier quoted context omitted.

Goals in the context of AI aren’t the type of thing you’re arguing against here. AI can absolutely have goals — sometimes in multiple senses at the same time, if they’re e.g. soccer AIs. Other times it might be a goal of “predict the next token” or “maximise score in Atari game”, but it’s still a goal, even without philosophical baggage about e.g. the purpose of life. Those goals aren’t necessarily best achieved by h…

AI does not have goals, it has Tasks. Tasks assigned by an operator. An AI cannot generate goals, since they are logically meaningless.

If you want to call them “tasks” you can, but the problem still exists, and AI can and do create sub-tasks (/goals) as part of whatever they were created to optimise for.

You might find it easier to just accept the jargon instead of insisting the word means something different to you.

Re: DeepMind: A Generalist Agent

#292

Post by barry-cotter »

galaxyLogic wrote: post #279
kromem wrote: post #260

Earlier quoted context omitted.

If anything, AGI seems to be the sole deus ex machina that can avert the inevitable tragedy we're on track for as a result of existing human misalignment. "Oh no, robots are going to try to kill us all" has to get in line behind "oh no, tyrants for life who are literally losing their minds are trying to measure dicks with nukes" and "oh no, oil companies are burning excess oil to mine Bitcoin as we approach climate c…

No matter how smart an AI gets it does not have the "proliferation instinct" that would make it want to enslave humans. It does not have the concept of "specism" of it having more value than anybody else. AI does not see the value in being alive. It is like some humans sadly commit suicide. But a machine wouldn't care. It will be "happy" to do its thing until somebody cuts off the power. And it does not even care whe…

> No matter how smart an AI gets it does not have the "proliferation instinct" that would make it want to enslave humans.

If it has a goal or goals surviving allows it to pursue those goals. Survival is a consequence of having other goals. Enslaving humans is unlikely. If you’re a super intelligent AI with inhuman goals there’s nothing humans can do for you that you value, just as ants can’t do anything humans value, but they are made of valuable raw materials.

> It does not have the concept of "specism" of it having more value than anybody else.

What is this value that you speak of? That sounds like an extremely complicated concept. Humans have very different conceptions of it. Why would something inhuman have your specific values?

Re: DeepMind: A Generalist Agent

#293

Post by rgavuliak »

atleta wrote: post #261
joe_the_user wrote: post #247

Earlier quoted context omitted.

We should think more about the Human alignment problem. Absolutely this The possibility of a thing being intentionally engineered by some humans to do things considered highly malevolent by other humans seems extremely likely and has actually been common through history. The possibility of a thing just randomly acquiring an intention humans don't like and then doing things humans don't like is pretty hypothetical and…

I wouldn't say the latter is hypothetical, or at least unlikely. We know from experience that complex systems tend to behave in unexpected ways. In other words, the complex systems we build usually end up having surprising failure modes, we don't get them right the first time. It's enough to think about basically any software written by anyone. But it's not just software. I've just watched a video on YT about nuclear…

>In other words, the complex systems we build usually end up having surprising failure modes, we don't get them right the first time. It's enough to think about basically any software written by anyone. But it's not just software.

That is true, but how often does a bug actually improve a system or make it inefficient? Isn't the unexpected usually a degradation to the system?

Re: DeepMind: A Generalist Agent

#294

Post by c1ccccc1 »

sinenomine wrote: post #211
justinpombrio wrote: post #208

Earlier quoted context omitted.

I am quite scared of human extinction in the face of AGI. I certainly didn't jump on it, though! I was gradually convinced by the arguments that Yudkowsky makes in "Rationality: from AI to Zombies" ( https://www.readthesequences.com/ ). Unfortunately they don't fit easily into an internet comment. Some of the points that stood out to me, though: - We are social animals, and take for granted that, all else being equal…

The human extinction due to would be "hard takeoff" of an AGI should be understood as a thought experiment, conceived in a specific age when the current connectionist paradigm wasn't yet mainstream. The AI crisis was expected to come from some kind of "hard universal algorithmic artificial intelligence", for example AIXItl undergoing a very specific process of runaway self-optimization. Current-generation systems aka…

The phrase "AGI owner" implies a person who can issue instructions and have the AGI do their bidding. Most likely there will never be any AGI owners, since no one knows how to program an AGI to follow instructions even given infinite computing power. It's not clear how connectionism / using gradient descent helps: No one knows how to write down a loss function for "following instructions" either. Until we find a solution for this, the first AI to not to be "obviously erroneously evil" won't be good. It will just be the first one that figured out that it should hide the fact that it's evil so the humans won't shut it off.

We humans have gotten too used to winning all the time against animals because of our intelligence. But when the other species is intelligent too, there's no guarantee that we win. We could easily be outcompeted and driven to extinction, as happens frequently in nature. We'd be Kasparov playing against Deep Blue: Fighting our hardest to survive, yet unable to think of a move that doesn't lead to checkmate.

Re: DeepMind: A Generalist Agent

#295

Post by brador »

ben_w wrote: post #291
brador wrote: post #283

Earlier quoted context omitted.

AI does not have goals, it has Tasks. Tasks assigned by an operator. An AI cannot generate goals, since they are logically meaningless.

If you want to call them “tasks” you can, but the problem still exists, and AI can and do create sub-tasks (/goals) as part of whatever they were created to optimise for. You might find it easier to just accept the jargon instead of insisting the word means something different to you.

Tasks are assigned, Goals are desired.

It is not simply semantics.

Re: DeepMind: A Generalist Agent

#296

Post by mdp2021 »

edouard-harris wrote: post #178
hans1729 wrote: post #5

I’m not sure how to word my excitement about the progress we see in AI research in the last years. If you haven’t read it, give Tim Urbans classic piece a slice of your attention: https://waitbutwhy.com/2015/01/artificial-intelligence-revol... It’s a very entertaining read from a couple of years ago (I think I’ve read it in 2017), and man, have things happened in the field since then. If feels like things truly start…

That Tim Urban piece is great. It's also an interesting time capsule in terms of which AI problems were and were not considered hard in 2015 (when the post was written). From the post: > Build a computer that can multiply two ten-digit numbers in a split second—incredibly easy. Build one that can look at a dog and answer whether it’s a dog or a cat—spectacularly difficult. Make AI that can beat any human in chess? Do…

> And as for recognizing a cat or a dog — that's a problem so trivial today

Last time I checked - though it's been a long while I could not check thoroughly owing to other commitments - "«recognizing»" there was "consistently successfully guessing", not "critically defining". It may be that the problem was solved in the latest years, I cannot exclude it - but I have not seen around in the brief "news checking" exercise the signals required for the solution.

The real deal is far from trivial.

A clock can tell the time but does not know it.

Re: DeepMind: A Generalist Agent

#297

Post by mdp2021 »

DavidSJ wrote: post #285
croddin wrote: post #198

Earlier quoted context omitted.

I think of it as System 1 vs System 2 thinking from 'Thinking, Fast and Slow' by Daniel Kahneman.[1] Deep learning is very good at things we can do without thinking, and is in some cases superhuman in those tasks because it can train on so much more data. If you look at the list of tasks in System 1 vs System 2, SOTA Deep learning can do almost everything in System 1 at human or superhuman levels, but not as many in…

System 2 could be close behind System 2 by building on System 1 but it isn't clear how long that will take. This is happening since a few months ago: Wei et al (2022). Chain of Thought Prompting Elicits Reasoning in Large Language Models. https://arxiv.org/abs/2201.11903

> a series of short sentences that mimic the reasoning process a person might have when responding to a question

Worried about the choice of the word 'mimic' - which as usual seems to retain the usual distance from foundational considerations.

Edit: nonetheless, the results of "Chain of Thought Prompting Elicits Reasoning in Large Language Models" are staggering and do seem, at least in appearance, go towards the foundationals.

Re: DeepMind: A Generalist Agent

#298

Post by sailingparrot »

thomashop wrote: post #29
2bitencryption wrote: post #10

given that the same model can both: 1. tell me about a cat (given a prompt such as "describe a cat to me") 2. recognize a cat in a photo, and describe the cat in the photo does the model understand that a cat that it sees in an image is related to a cat that it can describe in natural language? As in, are these two tasks (captioning an image and replying to a natural language prompt) so distinct that a "cat" in an im…

Definitely possible. OpenAI's CLIP model already embeds images and text into the same embedding space. I don't know exactly how this particular model works but it is creating cross modal relationships otherwise it would not have the capacity to be good at so many tasks.

> OpenAI's CLIP model already embeds images and text into the same embedding space.

Not really. Embeddings from images occupy a different region of the space than embeddings from text. A picture of cat and the text "cat" do not resolve to the same embedding.

This is why DALL-E has a model learning to translate CLIP text embeddings into CLIP image embeddings before decoding the embedding.

Re: DeepMind: A Generalist Agent

#299

Post by ben_w »

brador wrote: post #295
ben_w wrote: post #291

Earlier quoted context omitted.

If you want to call them “tasks” you can, but the problem still exists, and AI can and do create sub-tasks (/goals) as part of whatever they were created to optimise for. You might find it easier to just accept the jargon instead of insisting the word means something different to you.

Tasks are assigned, Goals are desired. It is not simply semantics.

Your left is my right, and with you definition “get laid” is a task from the point of view of evolution and a goal from the point of view of an organism.

It’s in much the same vein that it doesn’t matter if submarines “swim”, they still move through water under their own power; and it doesn’t matter if your definition of “sound” is the subjective experience or the pressure waves, a tree falling in a forest with nobody around to hear it will still make the air move.

If AI do or don’t have any subjective experience comparable to “consciousness” or “desire” is also useful to know, and in the absence of a dualistic soul it must in principle be as possible for a machine as for a human (“neither has that” is a logically acceptable answer), but I don’t even know if philosophy is advanced enough to suggest an actionable test for that at this point.

(That said, AI research does use the term “goal” for things the researchers want their AI to do. Domain specific use of words isn’t necessarily what outsiders want or expect the words to mean, as e.g. I frequently find when trying to ask physics questions).

Re: DeepMind: A Generalist Agent

#300

Post by igorkraw »

c1ccccc1 wrote: post #294
sinenomine wrote: post #211

Earlier quoted context omitted.

The human extinction due to would be "hard takeoff" of an AGI should be understood as a thought experiment, conceived in a specific age when the current connectionist paradigm wasn't yet mainstream. The AI crisis was expected to come from some kind of "hard universal algorithmic artificial intelligence", for example AIXItl undergoing a very specific process of runaway self-optimization. Current-generation systems aka…

The phrase "AGI owner" implies a person who can issue instructions and have the AGI do their bidding. Most likely there will never be any AGI owners, since no one knows how to program an AGI to follow instructions even given infinite computing power . It's not clear how connectionism / using gradient descent helps: No one knows how to write down a loss function for "following instructions" either. Until we find a sol…

All of this AGI risk stuff always hinges on the idea of us building an AGI, while nobody has any idea of how to get there. I need to finish my PhD first, but writing a proper takedown of the "arguments" bubbling out of the Hype machine is the first thing on my bucket list afterwards, with the TL;DR; being "just because you can imagine it, doesn't mean you can get there"
Post reply on HN