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

John Carmack: I’m going to work on artificial general intelligence

facebook.com

Post reply on HN
Print view View original

811–820 of 930 posts

Re: John Carmack: I’m going to work on artificial general intelligence

#811

Post by MR4D »

simonh wrote: post #596
Animats wrote: post #313

This is encouraging. If you're going to work on artificial general intelligence, a reasonable context in which to work on it is game NPCs. They have to operate in a world, interact with others, survive, and accomplish goals. Simulator technology is now good enough that you can do quite realistic worlds. Imagine The Sims, with a lot more internal smarts and real physics, as a base for work. Robotics has the same issue…

I don't think that approach is going to work. For any clearly bounded and delineated task, such as a game, the most optimal, lowest energy and lowest cost solution is not AGI but a custom tuned specialist solver. This is why I don't think Deep Blue or Alphago are paths towards AGI. They are just very highly advanced single-task solvers. Now Alphago and it's implementation framework are much more sophisticated than De…

> They are just very highly advanced single-task solvers.

What if AGI is just a combination of very highly advanced single-task solvers?

I happen to believe that it is an emergent behavior once the complexity gets high enough, so AGI might just be a (large?) collection of AlphaGo solvers connected to different inputs.

Re: John Carmack: I’m going to work on artificial general intelligence

#812

Post by tirewarehouse »

foobiekr wrote: post #786
jjoonathan wrote: post #757

Earlier quoted context omitted.

I don't see any reason to be so dismissive. Scientific pursuits have an extremely steep risk profile and are systematically underfunded because nobody knows how to capture the value that comes out of them. If someone wants to chase one on their own dime, we should celebrate that contribution to society rather than dwell on the fact that the median (even 99th percentile) outcome is that the project goes nowhere and th…

I’m not being dismissive at all; it’s an alternate read on what he wrote and honestly closer to what he seems to be saying than the people reading this as his intending to go full bore.

It may not be your intent, but you are coming off as very dismissive.

Re: John Carmack: I’m going to work on artificial general intelligence

#813

Post by mantap »

hackinthebochs wrote: post #717
wpietri wrote: post #401

Earlier quoted context omitted.

Could you list a few fields of research where "X amount of person-hours of research by typical researchers will solve the problem"? I can't think of one.

It's not about fields specifically, but about particular problems within fields. An example is neuroscience discovering the function of some unknown functional unit of the brain. We have all the conceptual machinery to solve the problem, we just need to fill in the details. On the other hand, the problem of consciousness doesn't even have the conceptual machinery in place such that more details will lead to the solut…

In mathematics the word used for these kinds of problems is "inaccessible", e.g. Reimann Hypothesis or (previously) Fermat'a Last Theorem. I don't know if Carmack's chances are as good as Wiles' were but certainly better than the average joe. It's also the case that AI is a substantially younger field (arguably it was only possible to correctly evaluate ideas since powerful GPUs were released this decade) and so the difficulty of open problems including AGI is not yet known.

Re: John Carmack: I’m going to work on artificial general intelligence

#814

Post by marvin »

aerovistae wrote: post #788
ufmace wrote: post #755

Earlier quoted context omitted.

I don't see Elon as a genius at any kind of engineering. Everything he's done there was pretty easily foreseeable as being physically possible. What he is remarkably good at is selecting daring and potentially market-changing business goals, and executing against them consistently and aggressively despite naysayers. It's easy to say that it's probably possible to land a orbital rocket first stage. But who would bet a…

If you think what he's done was easily foreseeable as possible, you haven't been paying much attention to headlines the past fifteen years.

It's only obvious in retrospect. Every step along the way, there has been thousands of people saying "this is impossible" or "this is theoretically possible, but it can't be engineered" or "this is possible in principle, but it will be so costly to develop that it doesn't make sense".

When AGI is developed, it will seem obvious in retrospect. Participating engineers will receive middle-brow dismissals saying that this was obviously practically possible, since after all the human brain operates according to the laws of physics.

Re: John Carmack: I’m going to work on artificial general intelligence

#815

Post by the_af »

seanwilson wrote: post #750
the_af wrote: post #740

Earlier quoted context omitted.

Agreed about the meaningful choices and dynamically generated reactions from NPC, but general AI is not needed for this in my opinion. You also have to consider whether the complaints of "many" players matter when publishing a game. A percentage of vocal players will complain no matter what . Yes, they will complain even if you somehow implement true AI!

> Agreed about the meaningful choices and dynamically generated reactions from NPC, but general AI is not needed for this in my opinion. Maybe, but it would be an impressive demonstration of AI that would be very different to what has shown for Go, Chess and StarCraft. I think a compelling AI written short story for example would be leagues ahead of what is required to write a convincing chatbot e.g. you need an over…

AI-generated "true" fiction seems like scifi to me.

Of course an hypothetical "Turing Test" of fiction-writing might be able to fool some people, and in an age where Netflix has been accused of producing content "by algorithm" this seems increasingly possible, but...

... what is "true" or "good" fiction is up for debate. In fact, it's a debate that can never be solved, because there is no right answer except what it feels to you, your friends and authors you respect.

But that said, I seriously doubt it would fool me, and I think it won't be within reach of an AI any time soon, or ever, not without creating an artificial human being from scratch. And maybe not even then, because how many real people can write compelling fiction anyway? :)

Re: John Carmack: I’m going to work on artificial general intelligence

#816

Post by tirewarehouse »

streetcat1 wrote: post #156
ausjke wrote: post #46

he is 49 this year, considered as one of the most genius programmer on earth. There are professors still doing real work at 90+ year old(yes, the UT professor goodenough for Nobel prize), John has a long way ahead, best luck!

Just out of curiosity, how do you define a genius programmer? (vs regular programmer).

He ushered in modern gaming, and everything he did in that domain was way ahead of the curve.

Re: John Carmack: I’m going to work on artificial general intelligence

#817

Post by visarga »

nathias wrote: post #599
visarga wrote: post #457

Earlier quoted context omitted.

Humans are not AGI's. We're specialised in human survival, not general intelligence. We're pretty limited in intelligence in many ways, actually, and the environment doesn't support generality. Without a proper challenge an agent would not become super intelligent. The cost of developing such an intelligence would conflict with the need to minimise energy for survival.

We are AGI, the general part comes from language and specialization of brains for language use.

No, if we were we could figure out the genetic code, or how a neural net makes its decisions. But we can't because, among other things, we have a limited working memory of 7-12 objects.

Programmers know how it is to live at the edge of the capacity of the mind to grasp the big picture. We always reinvent the wheel in the quest to make our code more grasp-able and debuggable. Why? Because it's often more complex than can be handled by the brain.

An AGI would not have such limitations. Our limitations emerged as a tradeoff between energy expenditure and ability to solve novel tasks. If we had a larger brain, or more complicated brain, we would require more resources to train. But resources are limited, we need to be smart while being scrappy.

For the record I don't think there is any general intelligence on our planet. A general intelligence would need access to all kinds of possible environments and problems. There is no such thing.

There's also the no free lunch theorem - it might not apply directly here, but it gives us a nice philosophical intuition about why AGI is impossible.

> We have dubbed the associated results NFL theorems because they demonstrate that if an algorithm performs well on a certain class of problems then it necessarily pays for that with degraded performance on the set of all remaining problems. [1]

[1] https://en.wikipedia.org/wiki/No_free_lunch_theorem

Another argument relies on the fact that words are imprecise tools in modelling reality. Language is itself a model and as all models, it's good at some tasks and bad at other tasks. There is no perfect language for all tasks. Even if we use language we are not automatically made 'generally intelligent'. We're specialised intelligences.

Re: John Carmack: I’m going to work on artificial general intelligence

#818

Post by 0-_-0 »

cf wrote: post #277

I expect John Carmack to follow a trajectory of someone like David Ha who with little previous background started to write very creative and thought provoking papers ( https://scholar.google.com/citations?user=J1j92GsxVUMC&hl=en ) It won't be AGI by most definitions but I bet it'll be pretty cool and I'm happy to have that.

Dammit, I had the hypernetworks idea recently but he already did it 4 years ago! Nothing new under the sun...

Re: John Carmack: I’m going to work on artificial general intelligence

#819

Post by PaulHoule »

quicknir wrote: post #799
fastball wrote: post #363

Earlier quoted context omitted.

The problem with the "field of researchers" is that most of us aren't geniuses. We're just plugging away at problems, like normal people. The opportunity for a genius is to come in, synthesize all existing information on the subject, and then come up with a novel approach to the whole thing. In some part, I think that is what Elon Musk has been able to do effectively. He comes into a field that already exists, reads…

I normally don't bother but this comment is so profoundly ridiculous I had to say something. Tenured ML professors at the top 100 or so universities in the world aren't "most of us". A very large chunk of these people are geniuses. Those jobs are incredibly hard to get, and most of these people are reading everything that is getting published, on an ongoing basis, and are outputting something novel, on an ongoing bas…

Granted.

But the academic activity is focused around the kind of activities that Kuhn calls "Normal Science".

That is, ML researchers mainly do competitions on the same data sets, trying to put up better numbers.

In some sense that keeps people honest, it also lowers the cost of creating training data, but it only teaches people how to do the same data set over and over again, not how to do a fresh one.

So a lot of this activity is meaningful in terms of the field, but not maybe not meaningful in terms of useful use.

I saw this happen in text retrieval; when I was trying to get my head around with why Google was better than prior search engines, I learned very little from looking at TREC, in fact people in the open literature were having a hard time getting PageRank to improve the performance of a search engine.

A big part of the problems was that the pre-Google (and a few years into the Google age) TREC tasks wouldn't recognize that Google was a better search engine because Google was not optimized around the TREC tasks, rather it was optimized around something different. If you are optimizing for something different, it may matter more what you are optimizing for rather than the specific technology you are using.

Later on I realized that TREC biases were leading to "artificial stupidity" in search engines. IBM Watson was famous for returning a probability score for Jeopardy answers, but linking the score of a search result to a probability is iffy at best with conventional search engines.

It turns out that the TREC tasks were specifically designed not to reward search engines that "know what they don't know" because they'd rather people build search engines that can dig deep into hard-to-find results, and not build ones that stick up their hand really high when they answer something that is dead easy.

Re: John Carmack: I’m going to work on artificial general intelligence

#820

Post by MrZongle2 »

I don't know if Carmack has a good chance at achieving this goal, but I certainly wish him the best.

There are certainly far worse applications of a sharp mind like his, and if this is where his passion has taken him then I'm sure he will be productive.

Post reply on HN