Superposition in quantum cavities

Sebastian Hassinger

Welcome to the new quantum era. I'm your host, Sebastian Hassinger. We're gonna listen to another interview I conducted at the American Physical Society's Global Summit back in April. I was able to record 10 episodes during the conference, thanks to the help of APS, and frankly, could have doubled that number given enough time. There were so many talented scientists around who I was desperate to talk with. Building that backlog of interviews allowed me to focus on the documentary we shot at the Helgeland twenty twenty five conference in June, organized by Yale University and the Max Planck Institute on the hundredth anniversary of Werner Heisenberg laying the foundation of the uncertainty principle on that island. The conference was fantastic, and we captured many fascinating discussions. We're in the process of packaging up the talks and panels for the organizers now, and we'll shift our focus to the documentary proper soon. I will be continuing to use the backlog of interviews and trying to get these out more frequently. Apologies for the sound quality of this intro. I'm actually on the road. I'm recording this on my iPhone using Ferrite, but I forgot my lavalier mic. So I'm just using the iPhone's mic. And to the very kind listener who had very nice words to say about the interviews and the interviewees I get who are I agree, excellent, but also pointed out sometimes audio quality is not the best. This recording aside, I promise I will be putting more effort into the postproduction to try to clean up levels and make it easier to hear. So patience, please, with us, but we're we're getting better slowly but surely. Today, though, we'll be speaking with a researcher who earned her PhD at Yale University and is now professor at the National University of Singapore, part of the incredible community there of physicists who allow that country to produce novel and important research in quantum physics and quantum computing at the volume of a much, much larger country. At NUS, doctor Yvonne Gao and her group work on superconducting devices for both scientific and technological motivations. I learned a lot from this conversation with doctor Gao, and I hope you'll enjoy it. Thank you for joining us, Yvonne.

Yvonne Gao

It's a pleasure to be here.

Sebastian Hassinger

And have you been enjoying the week so far?

Yvonne Gao

Yes. I'm having a really intense, but also a lovely time.

Sebastian Hassinger

It's my first time at APS Summit, and it intense is definitely the word I would 4,000 physicists in one place.

Yvonne Gao

Absolutely. And I think more more often than not, I think of it as chaotic. Yeah. But chaotic good. Good.

Sebastian Hassinger

Yes. Chaotic good. Definitely. So so you're here. You're presenting a session as well as as some of your students are presenting sessions.

Yvonne Gao

Yes.

Sebastian Hassinger

So you're an experimentalist. What is the what is the sort of the crux of what you're you're presenting of your work this week?

Yvonne Gao

Yeah. Thank you for asking about that. I presented on Monday, you know, first first day.

Sebastian Hassinger

Fully jet lagged.

Yvonne Gao

Good. Yes. But but I think the the excitement, it helped me stay awake for sure. So I talked about actually an overview of the experiments we do in Singapore in the team. They all they're all centered around the idea of using cavities, which are essentially harmonic oscillators

Sebastian Hassinger

Right.

Yvonne Gao

To probe interesting effects in physics and quantum applications. So I think it was quite a deliberate choice. I was thinking about perhaps going in-depth on one particular experiment versus having an overview of several of the recent works we've done. I opted for the later because I felt like I guess maybe it's a good way for us to showcase our So work to this what I ended up talking about is essentially a series of work that we did using very small systems. So these are one cavity, one cubit common systems. I guess you don't even hear about these numbers very often, right? Many zeros now.

Sebastian Hassinger

Everybody's competing over numbers of cubits. Absolutely.

Yvonne Gao

And we're very proud that we only use one cubit and one cavity. So this entire series of three experiments were all using the small small kind of hardware, and we tried to build in creative features and techniques from control and measurement perspectives to tease out interesting dynamics and features on the harmonic oscillator. So for instance, for one of them we probed the loss mechanisms of a cat stage that lives in our cavity. And we essentially looked at the very intuitive pictures of a particular representation of the cat and understood that if we perform some engineering of its distributions in this representation, we have then we actually have access to a fairly simple way without any special hardware engineering to make the cat intrinsically more robust or resilient towards the main losses in such cavity.

Sebastian Hassinger

Interesting. So it doesn't require design. It's more just the is the control of the cavity or the

Yvonne Gao

That's right. That's right. So our goal was to keep the hardware as standard as

Sebastian Hassinger

possible and

Yvonne Gao

try to think about what we can do to the state or to the elements in our system so that they can behave in a way that's interesting or that's not the most intuitive or standard manner. So in this example, it was about mitigation for loss. In the other example that I shared in my talk, it was about how we can use, again, just these highly non classical features of the cat to perform very sensitive detection of a small rotation of the state. So all of these are just using very much standard

Sebastian Hassinger

Clean vanilla qubits.

Yvonne Gao

You can say that, yes. Can't just use qubits that's already people can take our process and run it on whatever they already have in the in the lab, actually.

Sebastian Hassinger

But if if you're looking at a cat state, though, is that does that mean it's a cat cubit here?

Yvonne Gao

Yeah. That's that's a good question. So not all cats are Cubans. Our cats are plain cats. They're not Cupid cats. Okay.

Sebastian Hassinger

So the reason I say that

Yvonne Gao

is because we don't think about them as encoding logical information. Okay. Because if you wanna do that, you have to think about, you know, what kind of control you do. You have the gates. Okay. And how do they behave Right.

Sebastian Hassinger

So you have an implementation.

Yvonne Gao

Gates. We have gates.

Sebastian Hassinger

You do.

Yvonne Gao

Okay. Yeah. It's just that in these particular examples

Sebastian Hassinger

I see.

Yvonne Gao

We were using cats as a non Gaussian resource state

Sebastian Hassinger

Okay.

Yvonne Gao

As a quantum kind of embodiment of a quantum state I and its unique characteristics.

Sebastian Hassinger

So the wave function of the system in that cat Yeah.

Yvonne Gao

Okay. Maybe if we even strip it back, strip it down a little more, we're looking at a cat in its very fundamental form, which is the superposition quantum superposition stage.

Sebastian Hassinger

Right.

Yvonne Gao

And and it's the feature of that quantum superposition that we're looking for.

Sebastian Hassinger

Interesting.

Yvonne Gao

Yeah. It's just cat is the most iconic and, I guess, convenient way for us to access that

Sebastian Hassinger

Yes.

Yvonne Gao

That intrinsic superposition.

Sebastian Hassinger

So, I mean, is the goal of that work to improve qubit fidelity, or is it more fundamental on your your probing sort of what what more scientific value you can get out of the control of that cavity?

Yvonne Gao

Yeah. It's a little bit of both. Mhmm. I think so with all this this whole series of single cavity, single qubit experiments, we do want to focus a little more on the fundamental physical effects and just looking at quantum superposition, quantum entanglement from a different light. But all of this also do have fairly direct implications or potential applications in quantum computing as qubits or as, say, slightly protect more protected qubits. So the new dynamics we engineer could also be ways to implement gates. So we're not actively pursuing that in this set of works, but they're closely related.

Sebastian Hassinger

I mean, imagine anything that sheds more light into the cause of loss is going to inform hardware design, fabrication,

Yvonne Gao

Exactly.

Sebastian Hassinger

And design, control, all that sort

Yvonne Gao

of stuff. Absolutely. Yeah. So I think our angle is perhaps a little different from the Yeah. Full on, you know, making cubits and gates Right. Perspective. And that's, I think that's one thing that I really enjoyed being being an academic for. It's Right. Yeah. We were doing this more from a curiosity led Yes. Perspective. Yeah. But of course, keeping the big picture of applying to quantum computing

Sebastian Hassinger

Right.

Yvonne Gao

Using these harmonic oscillators at some point. Yeah. But the I guess the entry point is that we're curious about how we can tap into these dynamics and features of really at the fundamental

Sebastian Hassinger

Yeah.

Yvonne Gao

Quantum physics level and see what we can do with it or what we can manipulate in a way that comes across in a different fashion.

Sebastian Hassinger

Right. Right. That's really interesting. And in terms of sort of deeper scientific knowledge or understanding learning, what do you think what sort of is the the crux of the sort of question that's most pressing in your mind that you're trying to to answer?

Yvonne Gao

Oh, that's a that's a really interesting question. I think there it's little it's difficult to, like, narrow it down Yeah. To one thing. But I guess my starting point for continuing my research in circuit QED, especially using the bosonic element in circuit QED, is really because it's a very versatile platform. It's very engineerable. Is that that even a word? Sure. So we can we can kind of

Sebastian Hassinger

You can say it's plastic,

Yvonne Gao

I suppose. Yes. It's true. It can

Sebastian Hassinger

be very

Yvonne Gao

bespoke. Right.

Sebastian Hassinger

Right.

Yvonne Gao

So we can build in these features to really enhance or amplify certain things we want to study. So that kind of gave me the playground to think about these scientific questions. So if I have to go back to the specific questions themselves that I'm interested in, I think it really still is at the very fundamental level. So personally, think I still haven't fully grasped entanglement

Sebastian Hassinger

You're not alone.

Yvonne Gao

In this kind of very physical observable way. I understand the mathematics when I write it down, but when we make experiments of real entangled states in real hardware, How do we quantify it? How do we understand its features? What breaks that entanglement? So all of these things are still, I guess, still a little abstract in market. So we do have so we actually have a series, a set of two or three experiments that will look into this from both on a single harmonic oscillator kind of way as well as entanglement across several different harmonic oscillators. So I think we're using that this is an example of how we're using this very versatile platform to probe these interesting questions.

Sebastian Hassinger

Well, you used the word playground in it. That does sound like it's such an intrinsic problem with quantum physics in that it is so unobservable. Right? And and the act of actually observing also completely changes the system that you're looking at. Indeed. But visualizing it, making it real is really hard. So it makes sense to have this sort of playground, as you put it, where you can control all the elements and and, you know, viscerally experience the phenomena that you're studying, I guess.

Yvonne Gao

Yeah. Yeah. You're absolutely right. I think a lot of what we do is trying to find the most intuitive picture to capture what these abstract physical phenomenon actually look like in the lab.

Sebastian Hassinger

Well, the way you described sort of trying to get the most pure form of the cat state too. I mean, there's a sort of simplicity to that approach that I can I can imagine? It's almost like, you know, a craftsperson or an artisan honing something over and over again. You can actually increase value incremental or increase quality incrementally in the in the experimental result, I assume, by by having this sort of relatively less complex kind of operating environment that you're recreating the state over and over and over again.

Yvonne Gao

Yeah. Yeah. I think that's that's a good way to think about it because in in our hardware, you know, in any quantum hardware, really, there are a million different parameters that you can play with. So the fact that our system is quite controllable and you can tailor it allows us to essentially single out a few things to focus on. So we either optimize them or we change them to really probe the effect that we want to see. So that's quite handy.

Sebastian Hassinger

And you're fabricating these devices in your lab yourself?

Yvonne Gao

We do, we do. But fabrication difficult.

Sebastian Hassinger

I know. But that must be I I think that's a relatively new capability in US, I assume. Right? I mean, that that you're you're sort of establishing this

Yvonne Gao

Yeah.

Sebastian Hassinger

This fab capability.

Yvonne Gao

Indeed. We had to build up

Sebastian Hassinger

Yeah.

Yvonne Gao

This infrastructure Right. And knowledge quite gradually, you know, slowly slowly but steadily.

Sebastian Hassinger

Yeah.

Yvonne Gao

But what's really amazing is just over the years, you know, we we've actually if we if we look at the field in general, I feel like for a long time now we can make devices, like decent devices. But I think we've always known that they've not reached their limits yet. And just over the past few years, there's I think, really a large burst of activity on investigating how do we really push

Sebastian Hassinger

the

Yvonne Gao

performance of these devices. And from a material level, from a very fundamental material science and process integration kind of perspective. And it's really amazing to see that and see the resulting improvements we've seen just in the recent years. So we're not yet at the state of the art level, but we're keeping a lookout for all the knowledge that people have been reporting and trying to learn and keep up with the pace that we've When

Sebastian Hassinger

you think about the tolerances with which we're engineering these systems now, it is really mind blowing. Right? Mean, to build something that's capturing subatomic behaviors, like, in a consistent way Indeed. Indeed. Incredible.

Yvonne Gao

It is extremely difficult. So actually, something we actively do in the lab is that before we start an experiment, we always simulate the effects we are going to observe with sort of the very pessimistic system performance level just to make sure that whatever we're building is going to be able to serve the purpose Right. Revealing these interesting facts. And, you know, sometimes I joke with my students that we use the Perman's qubits, but we can still study interesting physics.

Sebastian Hassinger

You mentioned multi cavity devices coming sort of in the future, sort of planning for that. What would change in the behaviors that you're observing? Like what would the outcome of the experiment be in a multi cavity device in the future?

Yvonne Gao

Yeah, thanks for asking that actually. Because I'm very excited about this experiment for somewhat a personal reason as well. My PhD, I worked on this, one of my earlier projects was making a cat entangled across two cavities. And I think at that point it was sort of the first demonstration of this kind of mesoscopic entanglement across two distinct harmonic oscillators. So when I started my own team, I essentially wanted to go to three because one is single, two is a pair, and three is many. Yeah.

Sebastian Hassinger

Maybe better.

Yvonne Gao

Yeah. So I think with three we will have access to many much richer dynamics. So we planned this experiment where we're going to have still a single qubit but coupled to three harmonic oscillators. And we're trying to use this sort of more generalizable single qubit to multi cavity control process to create tripartite entanglement. Interesting. So this entanglement has been well studied in theory really and there are different classes of tripartite entanglement, and our device in principle gives us access to all of them. With that, our goal is really to see, to understand the structure of entanglement. There are very distinct, the different classes of them, so having access you know, the whole range through a single device allows us to, you know, characterize them in the same way and probe them and really understand their distinct By

Sebastian Hassinger

a single qubit. So there's it's a Josephson junction, it's a qubit, and then the three resonators are all linked to that that one junction. You were able to probe each of those resonators with microwaves Yes. Pulses separately.

Yvonne Gao

That's right. That's that's right. So we can do that both in a collective way as well as individually. So so it's quite I hope. It's it's ongoing right now.

Sebastian Hassinger

Sounds good.

Yvonne Gao

But but it is very exciting to be able to, you know, look into this really I think as three three resonator three cavities, we can start to call it really multipartite Yeah. Entanglement.

Sebastian Hassinger

Because it's in the resonator that those are still bosonic states that

Yvonne Gao

you're Yeah. So we're still at this very interesting boundary of, you know, we are using the base elements, our semiclassical states, but we're putting them in a highly entangled and nonclassical.

Sebastian Hassinger

How are the base elements semiclassical?

Yvonne Gao

Well, coherent states, which are, you know, what makes cats, essentially, they are essentially what we call semi classical states, that's because they are essentially something that you can almost back to just pendulums and how they behave under a force driving force.

Sebastian Hassinger

Right. Right. Okay. It's it's an oscillator, essentially. Right. Okay. Okay. Yeah. It's like

Yvonne Gao

the state a classical oscillator would have.

Sebastian Hassinger

Right. I see. But it's but it's also in a superposition, so that's why it's semi classical.

Yvonne Gao

Okay. No. The state itself is not in a superposition, but we can put them in a superposition

Sebastian Hassinger

I see.

Yvonne Gao

So that then the quantum feature comes out.

Sebastian Hassinger

I see. Interesting. That's really good. I mean, I can see how that kind of again, that's an expanded playground. There's more possibility for what you can do with it. I mean, do you think I guess there's both research value in that, but there's also pedagogical value. I mean, seems like it would be a very valuable teaching tool as well.

Yvonne Gao

I hope so. I haven't thought about it in that way. I think a lot of it is I think it'll be fun. Yeah.

Sebastian Hassinger

Well, that's sort of key to teaching, isn't it? Making it fun. No. Just back to what you were saying about, like, the difficulty of sort of grasping the the reality of these quantum phenomena, it sounds like the divide the multipartite entanglement would be a really powerful demonstration of of some of that quantum strangeness.

Yvonne Gao

You're right. Yes.

Sebastian Hassinger

Yes. It's really fascinating. So when you get up to multi cavity devices, is there any sort of thinking about sort of logical either operations or even things like error mitigation or error correction or control mechanisms of that multipartite entanglement or that multi cavity device?

Yvonne Gao

Yeah. That that's a good point. So we are not currently working on that, but there has been lot there's been lots of activity on that in the community. So there's actually a whole a couple sessions yesterday and today on multi cavity architectures, how they can be used for error correction codes, and the kind of control that you can have by being clever with your coupling element, etcetera. So that's definitely a very promising way to think about error correction and to even realize a different kind of error correction codes and protocols. So that's, I think, something that's not immediately on our But agenda right I guess going back to the earlier point is like what we're doing maybe more for curiosity and

Sebastian Hassinger

fun Right.

Yvonne Gao

Is certainly also very much complementary to what the rest of the community is working on in terms of making better logical qubits, better gates, and better processors with harmonic oscillators.

Sebastian Hassinger

Right. Right. Well, seems, as I said before, the the deeper we understand these phenomena, the better we can, you know, better informed the device design and fabrication can be. Do you think I mean, it it seems like there's sort of a unique opportunity right now where the, let's say, technological objectives, things like building more capable systems, are really closely aligned with the sort of scientific objectives. So, you know, curiosity driven research can interplay and have a direct impact on sort of the technological agenda. Is that that feels unique to me at this point in time. Is that something you

Yvonne Gao

I fully agree with you. And I do, I think, talk about this quite often with with my colleagues because I think it's very fortunate to be in this field at this time and that there's this nice synergy between the drive to make practical quantum processors and the, just like you said, the more academic and curiosity driven research focusing on the fundamental And scientific then there's this nice flow of knowledge both ways, really. What we do, what we learn, you tend to be perhaps a little higher risk sort of thing. And after we've done a proof of principle demonstrations, then that can be taken over, optimized, improved, and and perhaps becoming an ingredient for, you know, information processing and computing. But at the same time, a lot of our friends in the in the industry or in these large research efforts focusing on building processors, they develop such nice tools from design, from fabrication, for example, that they still publish, they share with the community just like what they're doing here, and that we can learn from that because we don't have the same resources of having a dedicated specialist team focusing on doing fabrication. So but through their learning, we're able to continuously upgrade and improve our own process as well. Right. So I think this kind of nice flow and this synergy is quite unique indeed to the field at this point. And I really enjoy it. I think it's a lovely time to be part of it at this point. And I do think it helped both sides Right. To thrive and go faster in whatever we're trying to do.

Sebastian Hassinger

Yeah. And I mean, you know, even the roots of the technological agenda are are sort of found in in Feynman's keynote at Endicott in 1981. If you wanna simulate nature, you

Yvonne Gao

use a

Sebastian Hassinger

quantum computer. So, I mean, in a way, if we build, you know, a a scaled a fault tolerant quantum computer at scale, essentially, that's almost like a virtual quantum physics lab. Mhmm. Right? You can imagine a perfect quantum computer could simulate any quantum behavior. If that did exist, what would be sort of your your top three wish list? And what would you wanna do with virtual a perfect virtual quantum lab?

Yvonne Gao

Oh, wow. That is that is a really good question that I have not been allowing myself to think about so much, which I should I should think about it more. I guess just off the top of my head, I will still really go back to the the basics. I want to be able to, you know, just see with my eyes all the textbook things that I've learned as a student and just, like, try it out on a real device Yeah. And then see exactly how they behave.

Sebastian Hassinger

Right.

Yvonne Gao

You know, this is as easy, as simple as say, just interfering many quantum elements

Sebastian Hassinger

Right.

Yvonne Gao

To more complex things. You know, we learn about forming bonds between two atoms and how this is all purely quantum mechanical. Right? Right. How they're, you know, in the quantum model of the atom is the overlap of their Right. Certain the wave functions of certain Like the original quantum application. Think it's binding Exactly. And we've learned that. I know how to calculate it. Yeah. But I think to be able to, like, you know, put that translate that on a on a real in a real physical thing and and see the outcome, I guess, more materially, more physically. That could be quite exciting. Really just translating a textbook to, like, a fun

Sebastian Hassinger

A simulated Yeah. That's terrific. Well, this has been excellent. Super interesting. Thank you so much for joining me.

Yvonne Gao

Thank you for the really nice conversation. My

Sebastian Hassinger

pleasure. I wish you the most luck in in future endeavor. I can't wait to see what you build in the future.

Yvonne Gao

So Thank you very much. I think we'll stick to one qubit for a little while. Yeah.

Sebastian Hassinger

Thank you.

Yvonne Gao

Thanks.

Sebastian Hassinger

Thank you for listening to another episode of the podcast, a production of the New Quantum Era hosted by me, Sebastian Hassinger, with theme music by OCH. You can find past episodes on www.newquantumera.com or on blue sky at newquantumera.com. If you enjoy the podcast, please subscribe and tell your quantum curious friends to give it a listen.