The fact that we don't have swarms of drones prefabbing houses in the relatively benign atmosphere of our planet should be evidence that maybe these claims aren't quite matched with reality.
There is also a very important step called randomization. Is that you have to take the same environment and then randomize the conditions. So, the cable doesn't literally only, you know, uh bend this way. It can bend a different way or the box can have different sizes, colors, different lids, and all that.
and I would probably preach the same thing like don't mess around with the logo and put it in a different color and put it on a crazy background until people know it so well that you can play with it
Now maybe this isn't completely out of the question but this would be quite challenging uh to do and that's because the calculus is a little bit different. We're not just running compute to optimize for a use case. We're actually running the robot in the real world and using the hardware and attempting the task in the real world.
we see that the across the board the single PIO like pre-trained PIO7 model matches or outperforms the fine-tuned specialists that were developed with reinforcement learning post-training for those downstream tasks.
without metadata prompting when you add lower quality data from 80% data to 100% data the performance actually decreases which is perhaps not too surprising because you're adding lowquality data to your data mixture whereas with the metadata prompting the performance actually increases when you add that lowquality data
and trying to fold two boxes together isn't useful data that will teach the model how to get better at the task. And so that would be kind of wasting a lot of time on the robot attempting to go down the wrong path for solving the problem. And so instead of spending a lot of time trying to do that task, what we'll do is we'll actually have a human intervene and show the robot what to do and how to recover from that situation.
kind of going back to this reliability question, we took this policy and we ran it not just once, but we ran it for 13 hours straight. Uh and we basically wanted to evaluate is this policy not only good at making a latte once, but can it do so reliably to the extent that it would be needed to be useful in the real world?
It will probably be economically shrewd to lose money on early robot models in anticipation that the data they gather will be worth more later in improving newer versions.
Given the high cost of errors, you need to have a very high level of safety and a very high level of confidence on day one before you deploy your first robot, before you drive your your first autonomous mile.
If you had asked me this question in the lead up to my 2019 book, Ultralearning, I would have said no. I was a strong advocate for doing hard problem solving and the associated mental effort it requires as a way to learn difficult skills.
You look under the underneath though at the code and it was just a horrible unmaintainable mess. But the fact that people could program the programs that they wanted was a a significant step forward as opposed to I'm going to write a thousandpage requirements document and then wait 8 years and not get what I want which was the alternative that we were offering at the at the time.
But right now I think it's sitting somewhere between half million and five million lines of code, somewhere in there. Probably more on the half million side right now and with the next drop of an Anthropic model, we're probably going to see it jump up to a few million lines.
The robot arm base case saves some hourly labor at the expense of adding more skilled labor, and can easily be negative if it needs reprogramming more than a few times per year.
Jazz is messy, and musicians seem to court disaster night after night. What can product leaders learn from how these artists approach their art? Barrett’s entertaining book formed the backbone for
So there's no ground truth. You can't have prior knowledge if you don't have ground truth because the prior knowledge is supposed to be a hint or an initial belief about what the truth is.
But the reason why it's not the most important thing for the kind of skills that you saw when you visited us, it at some level I think it comes back to Moravik's paradox. So Morovik's paradox is basically that it's like you know if you know one thing about if you want to know one thing about robotics it's like that's that's the thing. Morovik's paradox says that basically uh in AI the easy things are hard and the hard things are easy.
Like now, like basically learning is not for these systems is not just learning from raw actions. It's also learning from words eventually be learning from observing what people do from the kind of natural feedback that you receive when you're doing a job together with somebody else.
And I think we'll see the same thing with automation where uh basically robot plus human is much better than just human or just robot. Uh and and that just like makes total sense. It also makes it much easier to get all the technology bootstrapped because when it's robot plus human, now there's a lot more potential for the robot to like actually learn on the job, acquire new skills.
And when you make a mistake and correct it, well, first you you've achieved the task because you've corrected, but you've also gained knowledge that allows you to avoid that mistake in the future. With driving, because of the dynamics of how it's set up, it's very hard to make a mistake, correct it, and then learn from it because the mistakes themselves have significant ramifications.
Uh but it uh I think we kind of like know like roughly the puzzle pieces and it's something that we need to work on and I think if we work on it and we're a bit lucky and everything kind of goes as planned I think single digit is reasonable.
But it seems like you can understand it through passive observation, which is pretty surprising to me. And again, I think hints at something underlying about the nature of reality in my opinion, beyond just the cool videos that it generates.
Nobody wants to follow a joyless leader. If your presence signals, "Everything is fine. I've got this all figured out. Please don't show me anything messy," your team will take the hint.
They are not useful in industrial environments where, if a humanoid robot can do it, an industrial machine which is specialized in the task can do it even better.
But they said, "Look, in the end, actually the only way to actually get a real brain in the VAT is actually to have a brain in a body." And it could be a robot body, but you still need a brain in the body. So I don't think LLMs will get there because they can't. You really need to be embedded in a world, at least that's the E-four idea.
If you think about how we teach quantum mechanics to our students, the Copenhagen interpretation, it’s a God-awful mess. No one’s going to accuse that of being very beautiful. I’m a fan of the many-worlds interpretation of quantum mechanics, and that is very beautiful in the sense that fewer ingredients, just one equation, and it could cover everything in the world.
Because you think of how many electric motors are made in the world. There’s like tens of thousands, hundreds of thousands of electric motor designs. None of them were suitable for a humanoid robot, literally none.
Falling consumer oil demand looks much more like driving the Camry 6000 miles per year instead of 12,000 miles rather than a shiny new electric car in every garage or robot taxis swarming everywhere.
The original usability book by Steve Krug is the best starter for anyone looking to learn the basics of the subject. Nothing I've read helps crystallize as well the idea of your complicated user flow actually being a mess.
A korrent is a belief a person has stated in their own words: one
sentence stating the claim, backed by a quote and a source, kept at
korrents.com.
Under a name here, the quoted block is what they actually said.
The korrent beneath it is the claim those words support, in
korrents' wording — tap it to see the record, its source, and who
else holds it.
Nobody here wrote their own korrents. They are compiled from public
statements, and a person can change their mind, which is recorded too.
About the English under a post
Some people here publish in a language other than English. Where they
do, this site shows a machine translation beneath the post, in
this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the
serif above is exactly what the person published, and it is what to
quote them on. A translation can be wrong in ways that matter,
especially about tone.
Only the post's own words are translated. A quoted post, a linked
article and a belief on korrents.com
are left in their original language.