My jaw hit the floor when I watched an AI master one of the world's toughest physical games in just six hours
My jaw hit the floor when I watched an AI master one of the world's toughest physical games in just six hours
This AI plays Labyrinth better than you, or any other human
But this sounds exactly the sort of thing that machines are better at that people, so it just feels completely unsurprising that it was good at the task.
Turning multiple dials to manage speed and direction is not normally how humans interact with the world, so we can we pretty shit at it.
A basic motor is completely designed to turn like this.
This feels no different to the machine learning tools used to train on Mario a decade ago.
Right. Computers doing the shit that we don’t want to do for a living while giving us time to do things like paint cows that don’t have two heads.
Technology has removed a lot of time consuming or boring jobs, but it also made us spend our time in front of the computers. The idea from the start was that we could live our lives while computers do our tasks. But we ended up on social media or in front of computer games.
It's great for companies though, since now they make money both when we work and when we are off work. The attention economy is very real.
I'm not really surprised, the main challenge of that game is motor control, something any machine can do with more precision than a human
I agree but also disagree. It's true that machines are capable of fine motor control much more quickly and accurately than humans. But this by itself is often not enough.
This achievement should be somewhat surprising because of Moravec's paradox: the observation that, opposite to what early AI researchers expected, intelligence and reasoning skills are comparatively easy for a computer to simulate, while sensorimotor skills are in fact incredibly hard. Notice how, for example, chess engines started beating human players in the 90s or so, but we still don't have a robot that can do something as simple as pick raspberries (because surprise, for a machine picking a raspberry is actually hard as shit).
My eyes bursted out of my sockets when an AI was able to multiply 8 prime numbers faster than a human.
They're calling everything "AI" nowadays... this sort of learning algorithm is old as fuck, here's a 8yo example. The main differences between both situations is 1) some sensor(s) being used to "tell" the algorithm about the board state, and 2) the barebones robotic arms messing with the board.
I don't get what the issue is calling it AI?
Even if skipping completely the discussion about what is "intelligence", the expression "artificial intelligence" has been used as a label for so many different technologies that it has become practically useless. It includes things like decision trees in games (even if a lot of them boil down to simple if/then statements), generative models, even theoretical systems that would reason in a human-like way. And evolutionary models like the one in the OP and the one in my link.
So it's basically the 20s version of what "smart" was in the 90s/00s. Like this:
OK, I'm being cheeky and exaggerating it in the image macro, but it should give you an idea.
Exactly. Not to mention, why the fuck is it a surprise that a computer twisting the knobs “at superhuman speed” would be better at this game than humans. Like, no shit. We can’t compute how the degrees at which we’re turning the knobs affects the speed of the ball, can’t store that information for next time, and find the best way not making the same mistakes twice. Because…we’re human. We don’t have that finely tuned ability…because we’re not machines.
So…this isn’t “AI” despite the robot hands they put in the thumbnail and no shit a dedicated computer could master this game. I’m surprised it took six hours.
Additionally, this shit is really easy to compute. It's all Newtonian physics, and there are only two relevant equations here, both simple: d = at²/2 + vt and a = g*sin(θ). It's really easy for a computer to reach those formulas, cancelling the advantage that humans would have (insight and actual knowledge of the system).
Here is an alternative Piped link(s):
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source; check me out at GitHub.
I hope their jaw is alright
I cringed at the headline but just posted it as is and thought the article was kinda interesting.
It sure won't after he's gonna discover that his wife has chosen to leave him for her new AI driven dildo.
It is just a matter of time
You don't need AI to do that, seriously, such a buzzword where a relatively simple algorithm would suffice, don't tell me it's harder than double pendulums or those ball bouncing contraptions tech students make since a decade or more
Not needing AI isn't the point. The point is that AI can do it, and AI doesn't require a programmer to design and debug a bespoke algorithm to accomplish a task. It would take a human a lot longer than 6 hours to perfect an algorithm to do this.
Not needing AI isn’t the point. The point is that AI can do it, and AI doesn’t require a programmer to design and debug a bespoke algorithm to accomplish a task.
Maybe we should stop to call "AI" everything that is able to solve something by bruteforce.
A true AI, given the board and the rules, should have understood in less than a picoseconds that it need to avoid the holes, like a human does. What this AI did was simply to learn the rules, and a human is still faster in this (at this game at least).It would take a human a lot longer than 6 hours to perfect an algorithm to do this.
Man, the game has the solution drawn on it. A human perfect the algorithm in less than 6 seconds and it probably solve the game in way less than 6 hours. The point of the game is to follow, not to find, the path.
Yeah, you have a point here
Tell me more about this simple algorithm that does the same thing but isn't AI
Here, the video the article is talking about. Save you from reading the author's life story.
Save you from reading the author’s life story.
I can probably do that and still have time to spend in the washroom before the video is over. Some of us read fast.
Here is an alternative Piped link(s):
https://piped.video/zQMKfuWZRdA
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source; check me out at GitHub.
While the link is useful, the smug takedown is uncalled for. Humans relate way more through personal stories like this. Without the story, the video is not impressive at all, as most will have now idea how difficult this achievement is. There is also something to be said about adding some flourish and passion in the story, instead of coldly presenting facts.
It's just like those shitty recipe sites that tell you their grandma's life story for hours before giving the recipe. Get to the point, who cares about the anecdotes of some writer?
I don't want to connect with everyone always everywhere. It's just like small talk, which may be acceptable or even essential in some cultures, while considering rude and wasteful where I'm from.
Oh yeah? Can it tilt the board all the way to one corner, then pop the other corner and send the ball flying right to the end?
No, it's amateur at best.
That's addressed in the article actually. They had to program it so as not to cheat when they found it actually trying to cheat.
The true ability of AI/machine learning is to find and abuse all the loopholes and errors that exist in the training set.
"The only winning move is not to play" was simply WORP maximising its reward function.
This is pretty much what I'd expect AI to be best at.
It's cool but my question is (I did not see this addressed in the article nor video but might have missed it) did it learn to win the game in general terms or only this one example? I mean, if the layout of the board was changed, would it still solve it?
They don't discuss it here, but it's most likely a reinforcement model that operates different generations of learned behavior to decide if it's improving or not.
It would know that the ball going in the hole is "bad", and then try to avoid that happening. Each move that is "good' is then kept in a list of moves it should perform in the next generation of its plan to avoid the "bad" things. Loop -> fail -> logic build -> retry. After 6 hours, it has mapped a complete list of "good" moves to affect it's final outcome.
The answer your question: no, it would not be able to use what it learned here on a different map of the board. It's building reactions to events based on this one board, and bound by rules. You could use the ruleset with another board, but it would need to learn it all again just as a human would.
The thing about these models is less that they will work (it is assumed they eventually will through trial and error), but how efficiently they will work. The number of generational cycles and retries is usually the benchmark when dealing with reinforcement, but they don't discuss that data here either.
Yes, but that's kind of my point
We see it learn something with insane precision but most often it is almost an effect of over-training. It probably would require less time to learn another layout but it's not learning the general rules (can't go through walls, holes are bad, we want to get to X), it learns the specific layout. Each time a layout changes, it would have to re-learn it
It is impressive and enables automation in a lot of areas, but in the end it is still only machine learning, adapting weights to specific scenario
It did learn to use shortcuts to skip parts of the maze, and had to be told not to. Super interesting!
Yes, but that's only because a generation found some random, specific motion that scored better. Not because it analyzed that doing a skip should be possible
When the AI can solve one of these I'll be impressed:
Here is an alternative Piped link(s):
https://piped.video/UA33LOViUfw
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source; check me out at GitHub.
A blast from the past... Damn, now I have the urge to recover mine... from somewhere in the storage room... if it still exist...
Hey, I also had a toy like that! Cool!
Oh! That thing! Takes me back.
The only thing that is hard about this game is to control the board, which is the concept of it.
Not sure if it's more interesting that an AI taught itself the PID instructions in order to deftly move the ball around, or if it's more interesting if a human programs the PID instructions to move the ball around. Sounds like a lot of electricity was used doing it the first way.