5th HLF – Press conference with experts on the topic of Quantum Computing: Panel with Jay Gambetta and Chris Monroe
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Alternative Title |
| |
| Title of Series | ||
| Number of Parts | 49 | |
| Author | ||
| License | No Open Access License: German copyright law applies. This film may be used for your own use but it may not be distributed via the internet or passed on to external parties. | |
| Identifiers | 10.5446/40105 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
2
00:00
Internet forumQuantum computerForschungszentrum RossendorfPhysicsBit ratePhysical lawMusical ensembleQuantum mechanicsPhysical systemSoftware developerMeasurementNoise (electronics)MereologyQuantum computerCASE <Informatik>Category of beingLimit (category theory)BitMathematicsRevision controlMultiplication signInformationUncertainty principleMetrologieGroup actionFundamental theorem of algebraQuantum cryptographyBasis <Mathematik>Quantum entanglementAlgorithmQuicksortDifferent (Kate Ryan album)Computer animationXMLMeeting/Interview
03:10
Internet forumBit rateDrum memoryQuantum computerEncryptionTheoryState of matterInformation securityAnalogyQuantum entanglementMereologyCartesian coordinate systemWaveInterpreter (computing)DivisorComputerAlgorithmQuantum cryptographyPhysical systemParameter (computer programming)MeasurementQubitDistanceWave functionWeightView (database)BitMathematicsRandomizationGravitationClassical physicsQuantum mechanicsVideo gameCodeGoodness of fitNumberWater vaporMessage passingUniverse (mathematics)Sign (mathematics)Link (knot theory)Data structureSet (mathematics)Mechanism designBodenwelleZoom lensField (computer science)Nichtlineares GleichungssystemTerm (mathematics)Position operatorHidden variable theorySuite (music)Variable (mathematics)QuicksortTelecommunicationInformationMeeting/Interview
11:05
Drum memoryInternet forumSpecial unitary groupMetropolitan area networkEmailIntegrated development environmentDegree (graph theory)BitRight angleVapor barrierPhysical systemQuicksortInteractive televisionInterpreter (computing)WaveTheoryError messageValidity (statistics)Atomic numberMoment (mathematics)Formal verificationFundamental theorem of algebraGame controllerLogic gateField (computer science)Perspective (visual)FrequencyQubitQuantum computerWave functionRandom number generationPrice indexSingle-precision floating-point formatState observerSimulationAbstractionProbability distributionLimit (category theory)Multiplication signJunction (traffic)Open setType theorySuite (music)Black boxSoftware developerAnalytic continuationDisk read-and-write headTrailRevision controlComputing platformMeasurementPulse (signal processing)Point (geometry)Noise (electronics)VibrationMultiplicationInstance (computer science)Degrees of freedom (physics and chemistry)Key (cryptography)Standard deviationMathematicsMeeting/Interview
17:56
Internet forum3 (number)Drum memoryUniform resource locatorBit rateAirfoilPhysical systemQuantum entanglementAlgorithmMeasurement in quantum mechanicsAtomic clockBitInsertion lossGroup actionParameter (computer programming)Field (computer science)Process (computing)QuicksortNatural numberDifferent (Kate Ryan album)Goodness of fitQubitEmailRandom matrixLink (knot theory)CausalityComputing platformForm (programming)Atomic numberLaserMultiplication signTerm (mathematics)Limit (category theory)Quantum computerInequality (mathematics)Game controllerNoise (electronics)PhysicalismPerfect groupQuantum mechanicsSpiralDivision (mathematics)Standard deviationComputerIntegrated development environmentLogic gateCASE <Informatik>Similarity (geometry)Endliche ModelltheorieRoundness (object)Electronic program guideMultiplicationOffenes KommunikationssystemError messageDisk read-and-write headMoore's lawDirection (geometry)Message passingResultantWaveMeeting/Interview
24:46
Internet forumBit rateView (database)InformationQuicksortDifferent (Kate Ryan album)ResultantPhysical systemMatching (graph theory)Fundamental theorem of algebraMeasurementQuantum computerElectronic mailing listInteractive televisionAnalogyWordSet (mathematics)State observerIntegrated development environmentLevel (video gaming)Wave functionParameter (computer programming)Term (mathematics)Multiplication signMereologyInequality (mathematics)EvoluteCausalityNumberSuperposition principleTheoryQuantum entanglementMathematicsBitPhysicalismMetreLimit (category theory)Semantics (computer science)Interpreter (computing)Quantum mechanicsMusical ensembleDegrees of freedom (physics and chemistry)Standard deviationQuantum stateData conversionField (computer science)
30:01
VideoconferencingInternet forumBit rateMIDIMathematicsComputerFeedbackAtomic numberSoftware testingQuantum computerPhysicalismQuantum mechanicsMeasurement in quantum mechanicsRandomizationConnected spaceStudent's t-testVacuumNoise (electronics)Quantum entanglementRootStability theoryCodeComputer programmingParameter (computer programming)NumberGoodness of fitBitLaserTouchscreenRandom number generationPlastikkarteWebsiteFrequencyRight angleMusical ensembleView (database)Streaming mediaComputer animationMeeting/Interview
33:07
Internet forumJSON
Transcript: English(auto-generated)
00:23
I'd actually like to ask both of your thoughts. The development of science and technology seems to be very stimulated by the changes to what we can observe and the limits of measurement. I was wondering if both of you
00:41
could comment on how that's become apparent in quantum and particularly in marine science. Well, the fundamental limit of almost any measurement, any observable, is given by quantum mechanics. And as you probably know, there is a certain inherent noise in a quantum system.
01:00
If you measure something, it responds to it, so it reacts. And in many limits of metrology, that sets the noise floor. So quantum information is a little bit different. You wonder, how can you compute with all this noise?
01:20
And well, that is indeed a problem for devices to actually work. They have to be very well isolated. But the interesting part of a quantum computation happens without measurement. And so that so-called back action doesn't happen. And then at the end of the day, you make a measurement. You don't really care if it back acts.
01:42
You just want to extract information out. So it's often thought that there's a conflict between those. But that's certainly not the case. But if you can do a precision measurement at the fundamental noise floor, given by quantum mechanics, you probably have a system that is interesting for quantum computing as well.
02:01
It just happens to be. So the systems that we both work on are indeed, they fall in that category. They're very well isolated. And they can act in ways that they're not observed by us or anything. I think Chris said it very well. There's a reason why ions, which initially started
02:23
as for doing precise measurements, are now doing great quantum information experiments. Quantum and measurement has been linked for a very long time. The uncertainty principle says there's a fundamental limit of what we can actually measure.
02:41
The fact that in quantum cryptography, Eve doesn't know what basis was sent between the two parties acts as, gives us that protection. And this is part uncertainty. Measurement is fundamental to quantum. I agree for algorithms like Chris was describing. It's more down to the entanglement.
03:01
But this is fundamental place of measurement in quantum. And I think there always will be. I have a very basic question, if I may. So the way that quantum computing is often described
03:23
is that it looks at everything all at once. And I've heard quantum physicists say, yeah, kind of, but not really. Could you explain that in an ordinary person's terms? So to start, I really hate that analogy.
03:41
And what happens is we try to make an analogy for things that we don't really experience every day. So you're faced with this quantum theory. We get to see it in the lab. But it doesn't agree to our intuition. It's not something that you observe walking around.
04:00
So people like to say, oh, the reason why an algorithm works is it's like doing everything at once. I dislike that. I prefer to think of there's two things that's fundamental to quantum. The first is the measurement that was brought up previously. The second is entanglement and this idea that the whole is more complicated than the parts.
04:23
And for a quantum algorithm, to me, I have a view. It's an overgeneralization. But I think of it more as the first part of most quantum algorithms is you make a superposition. The second part of the quantum algorithm is you come in there and you apply the problem. As Scott said in his talk, this is like putting negative signs in there.
04:40
And those negative signs now make a system that I can no longer split apart into its parts. The whole is much more complicated. And then the creativity of most algorithms is coming up with some way that you can make it all interfere down to a small number of outcomes that you can easily measure. And so it's more that you exploit entanglement,
05:03
not that you do everything at once. Like a superposition, which is a state of every possible thing and it has all positive signs between each of them, is not entangled. It's not a complicated state to describe. I might add to that that what's
05:23
interesting about quantum mechanics, as an experimentalist, I don't tend to think too much about the philosophical issues of when you measure something, you affect it, or I do sometimes think about a quantum computer doing many things at once.
05:40
And there are many interpretations of what go on, but they all predict the same answers in the end. So in a sense, interpretations aren't science in a way. It's whatever you're comfortable with. That sounds a little wishy-washy, but I agree with Jay. Quantum mechanics is a wave theory, and there are lots of waves in everyday life.
06:01
We see water waves, sound waves, even light waves, and it is instructive to make analogies on how these waves evolve. But indeed, this entanglement issue is so weird. It's not just that you're computing everything at once. When you manipulate one of the qubits,
06:21
in a sense, you're automatically wired up to many more qubits without any physical wires. That's, again, maybe a wishy-washy analogy people don't like, but it's the lack of wires that makes it work. You can't possibly have exponentially many wires between classical bits. Quantum mechanics gives you them
06:40
by the structure of the theory. So interpretations, David Deutsch, who's one of the luminaries in the field, he says, well, it's clear that quantum computations take place in many universes. What else could it be? I mean, it's almost nothing else makes any sense.
07:04
So it's an interpretation, and actually, it works. If he's happy with it, good for him. Thank you. Sorry, I have a question about quantum cryptography. Because, as I understand, there will be ways
07:20
of sending messages where it becomes clear if somebody is eavesdropped on them. And I was just wondering, is that something that is inherent in any quantum cryptography? Like encryption method, or is that something extra? Because I don't quite understand how far along this idea has come.
07:42
Is it something that naturally would be part of quantum cryptography, or is it something that involves some extra steps or some extra effort? No, there are very mature technologies, mostly involving photons, because they can go over long distances and so forth,
08:01
which, by the way, are not so good for quantum computing. But it's good for quantum communication, sending photons. And indeed, you can show, in theory, that you should be able to detect if there's an eavesdropper. I'm not in that community so much, but I understand that the serious people in that community, they make their living
08:22
not trusting such devices. And it only works if you can know with a very high probability that you're sending exactly one photon. If you happen to send two by mistake, somebody could, in principle, take that second one, even if it's very rare. All you need is a little information, and you can start to crack things.
08:42
So proving a quantum link is secure is something that is much more difficult. And in fact, many communities don't really subscribe to quantum cryptography. They say the usual suitcase and handcuffs is good enough.
09:01
But on the flip side, you probably know one of the applications for quantum computing is quantum decryption. And this comes down to a math problem of factoring. So it is kind of interesting that quantum is involved in both sides of encryption and decryption. I think, again, Chris has said this pretty well.
09:20
So fundamentally, the quantum net randomness is what gives the prediction, okay? But again, you can write down an ideal theory. And when Charlie Bennett with John Smolin did this demonstration, I forget the year, I think it was 1983, apparently Charlie couldn't hear the coding that they put, but John could.
09:42
So that points out the sender and receiver can be compromised. So when you build an actual device, like Chris has said, it comes down to what's getting transmitted, what's happening at the sender and receiver. For the idolized theory, yes, the randomness of quantum is what gives it its security.
10:06
Can we just ask him because he mentioned multiple universes, how do you feel about that? Are you sure about that? Let's, when you come down to interpretations of quantum mechanics, you can have a lot of arguments.
10:20
It's better to do it with a drink in hand. Fundamentally, they are all based on the same set of equations, whether you prefer many universes, magical wave function collapses, or hidden variables, sorry, non-local hidden variables, it's up to you what your favorite is.
10:44
My comment was more about how I understand entanglement in an algorithm, interpretation of quantum mechanics. I'm not sure we'll ever see an end to that until someone finally links it to gravitational physics or something like that.
11:03
Maybe one way to move us towards a sort of concrete engineering problem, which is something I want to ask about, but on this question of interpretation of quantum mechanics, one of the things that strikes me as interesting about the development of quantum computing
11:20
over the course of decades is that this collapse of the wave function, which back in the early days was this, it's a sort of mathematically uncomfortable thing to do. You're just, okay, suddenly, and now we choose. And that was, I mean, it worked in a certain sense. And now, as you're trying from an engineering perspective to understand what you maintain coherence,
11:44
what decoherence is, I mean, are there any ways in which these, you do think about these sort of philosophical abstractions in a different way because of the engineering problem you're trying to solve? And it's possible that there aren't, but I just wondered. So, one of the ways I like to think
12:03
of what we try and do when we design systems is there's a conflict that we have. If I want to control a system, I want to open it up to its environment to allow me to easily do gates. Once I open something up to its environment, any vibrations, noise, can lead to that thing to decohere. We have a pretty well-defined theory
12:23
that explains the interactions between a system and its environment, how this leads to decoherence, how this leads to measurement. But when you're designing these systems, yes. You want to design, say, for our systems, we want to design the ability to read the cavity out at the frequency of the cavity, but not let photons out at the frequency of the qubit.
12:42
So, understanding how it interacts with its open quantum system is definitely, it's all the work that we have to do as we fight this thing I like to refer to as this quantum conflict. I want to be able to do things, but I don't want bad things to happen. And I want to open it up in just the right of way to do my control and not allow it
13:02
to interact with the environment. Same with the readout. I want to open it up to read it out, but have it nicely isolated. Yeah, the problem with all platforms, it's just a conundrum. But maybe getting to your question, theoretically, there are at least two ways that we use to describe decoherence.
13:21
And one of them is, as you said, it just snaps at a random time. So you can do a Monte Carlo simulation in seeding random numbers and have the system collapse a certain time then average over many realizations. Or you can do another way where everything smoothly decays continuously. And you get the same answer using both methods.
13:41
I mean, it's more mathematical methods. But they come with interpretations. I think that the wave function collapse interpretation is very close to what you might really see in a lab. You see photon clicks. You don't see half of a click. You see one or none. But mathematically, you can also describe it as just a smooth decay of that single photon at some point.
14:01
And you think of that decay as being a probability distribution. The math, in the end, looks quite different in the techniques used. One is more like diffusion, in a way, and the other one is more stochastic. But they both, we all use both methods. Again, it doesn't matter how you interpret it. You're gonna get the same answer in the end.
14:24
Technologically, what is the biggest barrier at the moment with the experiments that you're working on? Because people have mentioned, obviously, decoherence. People have mentioned the difficulty of, well, in the photon example, of having something that can reliably
14:41
emit single photons. What are the key physical challenges that you are? I might answer very generally to start with. Any platform, and certainly, for instance, in atoms and superconductors, there's a generic problem. When you build the system bigger,
15:02
it's more difficult for it to maintain its coherence. It just gets big. I don't wanna say it becomes self-aware and it measures itself. Yeah, that's an interpretation. But small systems are, individual atoms, for instance, are a single superconducting junction. It's almost purely quantum-mechanical, very coherent,
15:22
because it's a very simple system. It has only one degree of freedom, so you can keep track of it. But when you get lots of degrees of freedom, and I'm wandering into interpretation land, I mean, you can't deal with all those degrees of freedom, so we don't. We just say, well, there's decoherence going on. I think in all of the technology platforms,
15:43
we not only have to increase the qubit number, but at the same time, the gate performance has to increase, and that would seem like a recipe. I mean, it's very difficult to do that. We know that if you make a system bigger, it becomes even harder to limit decoherence.
16:02
I basically agree. Gate error is the most important thing that we have to do, not only just the gate error within the two qubits, showing that we don't influence all the neighbors in the neighboring systems. We've got very good at making two-qubit systems, doing really high-fidelity gates,
16:22
put it in a multi-qubit system. Those controls that I talked about before, they can start to influence those neighbors. If we really wanna be able to do these computations and do them in a short amount of time, we have to do gates in parallel. We have to do two-qubit gates and not worry about crosstalk, and just really understanding
16:40
how we make multi-qubit devices larger whilst maintaining the same fundamental two-qubit gate fidelity. And is it very clear to you when, I guess this is going back to the limits of observation sort of thing, is it very clear to you when things are going wrong and how they're going wrong?
17:04
Because I sort of have this picture in my head still of quantum computing as kind of the black box that you're saying you don't know what goes on until you read the end kind of thing. I mean, when you're trying to build the systems, how do you debug of which bits going wrong and which bits interfering? So we generally refer to that as verification and validation.
17:23
This field has been going around for a while. There are many techniques that you can use to sort of give you some indication of what's going on. Generally speaking, if it's a really large coupling of some type and you know it, the ions community, even NMR design pulses, which are sensitive to that era.
17:40
So you have, I didn't talk about it in my talk, but we have a whole suit of tools that we can just try to really isolate different things. There's no answer now, like the one answer, you just do this one method and then everything's great. You use your knowledge about your device, the tools that we're developing,
18:01
we're making more tools all the time. And we learn where things are going. And then once we learn what's wrong with that, we try and work out the physics of why that happened, work out how to remove it and proceed to the next. I like to think of it as a spiral. We just go around and around, hit the largest error, work out how to remove it, come back and fix.
18:20
Like chasing your tail. Hopefully the tail gets smaller each time. I'm writing for a very layman audience, so please excuse the simplicity of my question, but could you place us a little bit on the timeline
18:40
in terms of the successes we've already seen in quantum computing and what the future might look like? Well, I think maybe we can speak respectively on the two platforms we know best. They're quite different. I think from atomic physics,
19:01
these are qubits that are natural. They're individual atoms, they're all the same, so you can replicate them with perfection when you leave them alone. When they're left alone, they're perfect qubits, and you can scale that way. And in my community, all of the work is in controlling them from the outside.
19:20
It's the controls we have, and these are typically laser beams, microwave fields. These are the knobs that we open up these beautiful qubits to, and we have to really do a good job engineering those. I would say over the last, there's been steady improvements, steady but slow, over the last 20 or 30 years
19:41
in this field. The good thing about quantum systems is that every time you add one qubit, in a sense, you double the size of the system. So if you just add a qubit every couple years, it'll do it each year. Thank you. That's like its own Moore's law. But the superconducting platform has different challenges. Yeah, so I think even coming back for the challenges,
20:05
I think what I said in my talk, I do think is true. There was a lot of foundational experiments, like really showing inequalities of quantum mechanics, showing two-qubit gates. The ions and NMR systems led the way. Superconducting qubits really started to emerge
20:21
in the early 2000s. And then we worked out how to really isolate this system, get it better and better. We still have a long way to go because these man-made atoms, they're not as good as what nature gives us. And we're making them, they're quite big, and we're making them more and more isolated, understanding where the loss goes. But the same challenges we have ahead is,
20:42
like both Chris and I are saying the same thing, as you make these multi-qubits bigger, you open up more paths for it to interact, you gotta understand what they are, and you gotta try and get rid of them. Whether it's lasers or whether it's waveguides that we bring in, different physics, similar problems, but you really gotta understand this system
21:01
and its open, what we call an open system, how it interacts with its environment. And I think what you're gonna see is, you're gonna see larger devices, both of the ions and superconducting qubits. And are you gonna see them in the hands of people that are not necessarily always looking at them, and they'll look at different techniques.
21:20
I'm quite excited that with the system that we did, people have already started to run and did their own characterisation. A group from Sandia took our device and said, you know what, you have crosstalk in your device. That's a good thing, like show me where it is, go learn, bring your own ideas. So I think what you're gonna see
21:40
is a community of people that are gonna start using these devices, because we've got them pretty good, we have a long way to go, and there's so many different things we can explore. I'd like to ask another form of that question in a slightly focused way. Not in a very direct, causal sense,
22:03
but loosely, Shor's 1994 result, that, okay, here's an algorithm for factoring things. How has that, if at all, shaped the field, in terms of having a signpost of like, oh, here's a thing that was somewhat non-obvious,
22:22
that gives a direction, and has it? Yeah, sure. Scott talked about showing the difference between classical and quantum. Shor's was the first example of something that, you know, if you build this thing, the best we know about either one thing is wrong. Quantum mechanics is wrong, which we strongly believe not to be the case.
22:42
Our understanding of what we can do classically is wrong, or there is something more powerful about this device, this new model of computation. So it sets as a good sort of thing for the field to say, it's more than what we can do classically. And either one of those three answers
23:00
would be a good thing. I don't think it's gonna be quantum mechanics being wrong. But this is Scott's argument. I'm just paraphrasing it. I guess what I'm asking, was that a surprising moment? I mean, there were people working on- That was before I was in the field, so I can't really say on that one. I like the algorithm. Yeah, I was not in the field.
23:21
Well, I was working with atoms. We were making atomic clocks. In fact, we were making atomic clocks a little better by entangling them, entangling the atoms. You can get better signal-to-noise. Again, the link between using entangled states and beating the limits of quantum measurement because the noise is correlated in a certain way. So, you know, it sounds esoteric,
23:41
but I was working with David Wineland in Boulder, Colorado at the standards lab there, the atomic clock division. It was very cool research because we were basically making quantum gates. We didn't call it that. And then when Shor's algorithm came along, I mean, it just landed in our laps. And literally government agencies visited us and said,
24:02
boy, would you mind if we gave you lots and lots of money for you to do the same thing you've been doing? And in exchange, back then it was easy. In exchange for a one-page email every year? Man, those are good times. So I was finishing high school and I got into quantum.
24:20
And so I got in. I will say when I read about Shor's algorithm, it was like, oh, this is cool. This is what I wanted to do. So a lot of that was just already worked out before I got into the field. Can I?
24:45
I'm sorry to ask, but I think we've both perhaps said you don't really like thinking about the philosophical side. But I quite like it when people who work with the experiments give you feedback on it. So one of the things we're trying to write something about is the role of the observer in physics over time.
25:04
And quantum's actually obviously fundamentally changed what the notion of quantum is. The notion of an observer is in physics. Have you got any sort of thoughts on either as your own experiences, experimentalists, or sort of your sort of thoughts in general? Oh, I mean, I don't know.
25:20
I don't want to speak for Jay, but I love reading popular quantum books. There's so many of them in there. Many of them are just very well-written analogies. In terms of the observer, I mean, we take data all the time that collapses the superposition. And again, I tend just not to worry too much.
25:41
I believe the wave function collapse interpretation, I just subscribe to it because it's easy. It's easy at least to grasp, even if you won't understand it. So I guess I don't lose too much sleep over it, but you can really, especially over a drink, you can really get caught in saying,
26:01
well, how come I'm not in a superposition of having observed A and B? How come, I mean, it comes down to maybe understanding what the brain is, I guess. And then how can the brain understand itself? So yeah, you tend to, it's a slippery slope to philosophy, it's really cool, actually. And what I love about it is like this room.
26:23
People love the field. It has an era of mysticism about it. And you know, not that we take advantage of it, but it's neat, it's cool cocktail conversation. So I think it'd be important in the thing you're writing about, and Chris touched it, is to really point out this sort of iron started
26:44
originally to make observations better. People were trying to really understand what is the fundamental way of measuring things, using entanglement to do that. And like understanding that is where a lot of this
27:00
quantumness has grown out of. So it is fundamental to the path. Yes, we've accepted it more now because we've been working with these systems for a long time. But at the start, you guys were trying to reach the standard quantum limit, get to the Heisenberg quantum limit, really understand this fundamental interaction between measurement and back action.
27:23
As for the interpretations, for a few years during my PhD, I did a bit on that, but I kind of just in the end accepted that the math is all the same. And I'm not sure which one I prefer. I like to think more as the system interacts
27:41
with an environment. That environment then, due to interacting with many more degrees of freedom, eventually the information starts to decohere. And what we need to do is make our system when we measure it in a way that, so we talked a bit about this collapse before.
28:02
You can think of, if I know how I'm measuring it or my meter that's measuring my system knows exactly that interaction, I can describe that evolution with a pure state. It's just stochastic. I don't know. I think I even have a paper with my own PhD supervisor that proposed a set of inequalities.
28:22
I know I have a paper that posed a set of inequalities that you could try and look at to see if there's a difference. But the efficiency of the measurement is not there for these systems. Jordan Bell had a list of words that he wanted to see expelled from quantum theory, didn't he? And among them was measurement and observer.
28:42
So he was one who felt that you don't have a full quantum theory until we get rid of those issues. But it doesn't seem clear that that's gonna happen. Well, it doesn't sound, I mean, that sounds more like a semantic argument. It doesn't sound like science somehow.
29:01
You know, what term you use and so forth. He was a smart guy, though. He has this book, Speakable and Unspeakable in Quantum Physics, a very good book. Sorry, we have time for one more question and then we're gonna have to wrap it up.
29:21
Well, come on. I had one question you were talking about when you were working on the atomic clock and that you were getting close to the Heisenberg limit. And what was it like, you know, cause that was a very obviously challenging problem but also that was one of the few small number
29:41
of people actually trying to take that on. I mean, what was it like to be involved in work that was sort of going to such a high level of precision that hadn't been done before? You know, a part of it is disappointing in that we always, back then we always knew what the result of any experiment was going to be
30:00
and if it didn't match what we thought it was gonna be, something was wrong in the lab. And so in fact, the signals we saw, they were only two atoms. We were basically looking for a square root of two gain in some parameter and of course we saw it and we expected it. What was cool though is all the toys that we, this involves pretty fancy lasers,
30:22
manipulating individual atoms that are nearly at rest in a vacuum chamber, pretty exotic stuff and getting all that, you know, under wraps and all working at once. So I'm not answering the question maybe as you want it. It was more of a technological marvel. The physics in the end of the day, I tell my students this, that 99% of the stuff you do
30:43
is just low noise circuitry, feedback loops, making things stable and so forth. The last 1% is the quantum measurement where it's all the cool stuff, but make no mistake, you're not gonna really learn much about quantum mechanics here. You're gonna learn how to make a classical system very quiet.
31:04
Can I speak in one quick word? I'm sorry, it may have been, I'm sorry I was late. Maybe it was answered. So in the earlier talk today, there was some question about what the hello world program might look like in quantum computing. Microsoft said they wrote one on Monday.
31:21
I don't know if you've seen that, which is a... I think that, I didn't look too much, but it looked like they were doing a teleportation circuit. I don't know what the answer is. We, I like the Bell State, Bell State's similar to teleportation, but something, maybe I'll throw it back to you.
31:43
What would you guys like to see as a hello world? I feel like hello world prints something to my screen, right? And then I feel a connection. Oh, I did some code and then something came. So you need something to come back. I don't know what the answer is. I was more throwing it up as now you can start to program these things.
32:01
What do people think the hello world is? Maybe random generating quantum random numbers, generating entanglement. Teleportation is a good example. There's a quantum number generator at ANU, which is there. ANU's most visited website, of all their websites, Australian National University.
32:21
They have a quantum number generator. Well, it's subtle. I mean, you can never prove bits, you can never prove numbers are random, period, because somebody else could have a copy. It might look random to you, but using quantum mechanics and entanglement, you can prove that you have private randomness, even though the bitstream, there's no tests that will show
32:40
that one bitstream is perfectly random. You can show that nobody can have a copy. It's kind of cool. Thank you. Thank you. Thank you very much.
Recommendations
Series of 38 media