So, where are the robots? We've been told for 40 years already that they're coming soon. Very soon they'll be doing everything for us. They'll be cooking, cleaning, buying things, shopping, building. But they aren't here. Meanwhile, we have illegal immigrants doing all the work, but we don't have any robots. So what can we do about that? What can we say? So I want to give a little bit of a different perspective of how we can perhaps look at these things in a little bit of a different way. And this is an x-ray picture of a real beetle, and a Swiss watch, back from '88. You look at that -- what was true then is certainly true today. We can still make the pieces. We can make the right pieces. We can make the circuitry of the right computational power, but we can't actually put them together to make something that will actually work and be as adaptive as these systems.
Pa, gde su roboti? Već 40 godina nam govore da roboti dolaze uskoro. Uskoro će raditi sve za nas. Kuvaće, čistiće, ići u kupovinu, gradiće. Ali nisu ovde. Sa druge strane, ilegalni imigranti rade sav posao, ali nemamo robote. Šta možemo uraditi povodom toga? Šta možemo reći? Želim da pružim drugačiju perspektivu kako bismo mogli da gledamo na ove stvari na drugačiji način. Ovo je rendgenski snimak prave bube i švajcarskog sata, iz 1988. Pogledajte to - ono što je tad bilo istina i danas je. Još uvek možemo proizvesti delove. Prave delove. Možemo napraviti kola prave računarske snage, ali ne možemo ih zapravo spojiti da bismo napravili nešto što će zaista raditi i biti prilagodljivo poput ovih sistema.
So let's try to look at it from a different perspective. Let's summon the best designer, the mother of all designers. Let's see what evolution can do for us. So we threw in -- we created a primordial soup with lots of pieces of robots -- with bars, with motors, with neurons. Put them all together, and put all this under kind of natural selection, under mutation, and rewarded things for how well they can move forward. A very simple task, and it's interesting to see what kind of things came out of that.
Hajde da probamo da gledamo iz drugačije perspektive. Hajde da pozovemo najboljeg dizajnera, majku svih dizajnera. Hajde da vidimo šta evolucija može da uradi za nas. Napravili smo prvobitnu supu sa puno delova robota - sa polugama, motorima i neuronima. Sastavite ih zajedno i sve ovo prepustite mutaciji kao prirodnoj selekciji i nagradite stvari na osnovu toga koliko brzo se kreću unapred. Jednostavan zadatak i zanimljivo je videti šta je sve nastalo iz toga.
So if you look, you can see a lot of different machines come out of this. They all move around. They all crawl in different ways, and you can see on the right, that we actually made a couple of these things, and they work in reality. These are not very fantastic robots, but they evolved to do exactly what we reward them for:
Ako pogledate, možete videti da je dosta različitih mašina nastalo iz ovoga. Sve se kreću. Pužu na različite načine i sa desne strane možete videti da smo zapravo napravili par ovih stvari i da rade u stvarnosti. Ovo nisu naročito fantastični roboti, ali su se razvili da rade tačno onako kako ih nagrađujemo:
for moving forward. So that was all done in simulation, but we can also do that on a real machine. Here's a physical robot that we actually have a population of brains, competing, or evolving on the machine. It's like a rodeo show. They all get a ride on the machine, and they get rewarded for how fast or how far they can make the machine move forward. And you can see these robots are not ready to take over the world yet, but they gradually learn how to move forward, and they do this autonomously.
za kretanje unapred. To je urađeno u simulaciji, ali možemo to uraditi i na pravoj mašini. Ovo je fizički robot kod kog zapravo imamo populaciju mozgova, koji se takmiče ili se razvijaju na mašini. To je kao rodeo šou. Svi mogu da jašu na mašini i nagrađuju se za to koliko brzo ili daleko mogu da pomere mašinu unapred. Vidite da ovi roboti još uvek nisu spremni da preuzmu svet, ali postepeno uče kako da se kreću unapred i ovo rade samostalno.
So in these two examples, we had basically machines that learned how to walk in simulation, and also machines that learned how to walk in reality. But I want to show you a different approach, and this is this robot over here, which has four legs. It has eight motors, four on the knees and four on the hip. It has also two tilt sensors that tell the machine which way it's tilting.
Na ova dva primera, u principu smo imali mašine koje su u simulaciji naučile da hodaju i mašine koje su naučile da hodaju u stvarnosti. Ali želim da vam pokažem drugačiji pristup, na ovom robotu, koji ima četiri noge. Ima osam motora, četiri na kolenima i četiri na kukovima. Takođe ima dva senzora za nakretanje koji mu govore na koju stranu se nakreće.
But this machine doesn't know what it looks like. You look at it and you see it has four legs, the machine doesn't know if it's a snake, if it's a tree, it doesn't have any idea what it looks like, but it's going to try to find that out. Initially, it does some random motion, and then it tries to figure out what it might look like. And you're seeing a lot of things passing through its minds, a lot of self-models that try to explain the relationship between actuation and sensing. It then tries to do a second action that creates the most disagreement among predictions of these alternative models, like a scientist in a lab. Then it does that and tries to explain that, and prune out its self-models.
Ali ova mašina ne zna kako to izgleda. Vi je gledate i vidite da ima četiri noge, mašina ne zna da li je zmija ili drvo, nema predstavu kako izgleda, ali pokušaće da to sazna. Isprva napravi nasumični pokret i onda pokušava da shvati kako bi mogao da izgleda. Vidite puno stvari koje im proleću kroz um, puno samostalnih modela koji pokušavaju da objasne vezu između aktiviranja i detekcije. Onda pokušava da uradi drugu radnju koja stvara najviše nesporazuma između predviđanja ovih alternativnih modela, poput naučnika u laboratoriji. Onda uradi to i pokušava da objasni tu radnju i izdvoji model samog sebe.
This is the last cycle, and you can see it's pretty much figured out what its self looks like. And once it has a self-model, it can use that to derive a pattern of locomotion. So what you're seeing here are a couple of machines -- a pattern of locomotion. We were hoping that it wass going to have a kind of evil, spidery walk, but instead it created this pretty lame way of moving forward.
Ovo je poslednji ciklus i kao što možete videti otprilike je shvatila kako izgleda. A kada ima model samog sebe, može da ga iskoristi da smisli šablon kretanja. Ovde vidite nekoliko mašina - šablon kretanja. Nadali smo se da će imati nekakav zloban hod, poput pauka, ali je umesto toga smislila ovaj prilično dosadan način kretanja unapred.
But when you look at that, you have to remember that this machine did not do any physical trials on how to move forward, nor did it have a model of itself. It kind of figured out what it looks like, and how to move forward, and then actually tried that out. (Applause)
Ali kad to pogledate, morate se setiti da ova mašina nije fizički pokušala da se kreće unapred i nije imala model svog izgleda. Otprilike je shvatila kako izgleda i kako da se kreće i onda je to zapravo isprobala. (Aplauz)
So, we'll move forward to a different idea. So that was what happened when we had a couple of -- that's what happened when you had a couple of -- OK, OK, OK -- (Laughter) -- they don't like each other. So there's a different robot. That's what happened when the robots actually are rewarded for doing something. What happens if you don't reward them for anything, you just throw them in?
Krenućemo ka drugačijoj ideji. To se desilo kada smo imali par - to se desilo kada ste imali par - OK, OK, OK - (Smeh) - ne vole jedan drugog. Dakle, to je drugačiji robot. To se desi kada robote zaista nagradite za nešto što rade. Šta se desi kada ih ne nagradite i samo ubacite unutra?
So we have these cubes, like the diagram showed here. The cube can swivel, or flip on its side, and we just throw 1,000 of these cubes into a soup -- this is in simulation --and don't reward them for anything, we just let them flip. We pump energy into this and see what happens in a couple of mutations. So, initially nothing happens, they're just flipping around there. But after a very short while, you can see these blue things on the right there begin to take over.
Imamo ove kocke, kao što vidite na dijagramu. Kocka može da se rotira ili okrene na stranu, samo smo ubacili 1 000 ovih kocki u supu - ovo je u simulaciji - i nismo ih nagradili ni za šta, samo smo ih pustili da se okreću. U ovo pumpamo energiju i gledamo šta će se desiti kroz nekoliko mutacija. U početku se ne dešava ništa, samo se okreću u krug. Ali nakon kratkog vremena, možete videti da ove plave stvari sa desne strane počinju da preuzimaju vođstvo.
They begin to self-replicate. So in absence of any reward, the intrinsic reward is self-replication. And we've actually built a couple of these, and this is part of a larger robot made out of these cubes. It's an accelerated view, where you can see the robot actually carrying out some of its replication process. So you're feeding it with more material -- cubes in this case -- and more energy, and it can make another robot. So of course, this is a very crude machine, but we're working on a micro-scale version of these, and hopefully the cubes will be like a powder that you pour in.
Počinju da se međusobno kopiraju. U nedostatku nagrade, suštinska korist je kopiranje samog sebe. Zapravo smo napravili par ovih robota i ovo je deo većeg robota napravljenog od ovih kocki. Ovo je ubrzan snimak gde možete videti da robot izvršava deo procesa kopiranja. Ako ih hranite sa više materijala - u ovom slučaju kockama - i sa više energije, mogu da naprave još jednog robota. Naravno ovo je veoma sirova mašina, ali radimo na verziji mikroskopskih dimenzija, s nadom da će kocke biti poput praška koji se može usuti
OK, so what can we learn? These robots are of course not very useful in themselves, but they might teach us something about how we can build better robots, and perhaps how humans, animals, create self-models and learn. And one of the things that I think is important is that we have to get away from this idea of designing the machines manually, but actually let them evolve and learn, like children, and perhaps that's the way we'll get there. Thank you. (Applause)
Ok, šta možemo naučiti? Naravno, ovi roboti sami po sebi nisu veoma korisni, ali mogu nas naučiti nešto o tome kako napraviti bolje robote i možda o tome kako ljudi i životinje prave modele sami sebe i uče. Jedna od bitnih stvari je da moramo da zaboravimo na to da dizajniramo mašine ručno i da ih pustimo da se same razvijaju i uče, poput dece, i možda ćemo na taj način doći do toga. Hvala vam. (Aplauz)