"Building a quantum computer is not a solo task"
Solving one engineering problem at a time
Solving complex quantum computing problems and moving halfway around the world to do so – Emily Wright views her Ph.D. as one big adventure. Always alternating between theory and experiment, she is researching the optimization of quantum gates at the Walther Meissner Institute and has never lost sight of how exciting and almost magical she finds quantum computers.
By Veronika Früh
When Emily Wright moved from Canada to Munich eight months ago, there was one essential she needed to bring: "My hockey gear!", she exclaims with a laugh. When she first got in touch with her new colleagues-to-be at the Walther Meißner Institute (WMI) and asked for help at her arrival, she told them she would be easy to recognize with her hockey stick and hockey bag. "They thought I was kidding with them, because I am Canadian." Eight months later, Emily has fully settled in in Munich and is pursuing her Ph.D. at the WMI, working on optimal control for superconducting quantum circuits.
The first time Emily came into contact with the field of quantum computing might be a bit unusual. During an internship at the Bank of Canada, she attended a talk by the quantum hardware company D-Wave. "It was the first time that I heard of the idea of quantum information and how we can have this new way of thinking about information and new ways of doing computation," she recounts. "And I was really fascinated by that." So fascinated, in fact, that after her Bachelor’s degree in Applied Mathematics and Engineering at the Queens University in Kingston, Canada, she decided to switch it up a bit and go for a Master’s degree in Physics at the University of Victoria. "When I got interested in quantum computing, I wanted to understand a little bit more of the science behind it and how it worked, the physical implementations" the 24-year-old recounts. The fact that it is still such an open field, with so many engineering challenges left in it – some of them for her to solve, as she enthuses – is what really intrigued her.
Designing fast quantum gates
One of Emily’s main projects in her Ph.D. is to design fast quantum gates with deep reinforcement learning. "I can break that down a little, because there are quite a few different terms in there," she says with a big smile. Her first career aspiration – becoming a teacher – really shines through as she gets deeper into the explanation. "I like being able to take complicated problems and turn them into something that is more simple, solvable and understandable to other people," she adds.
Her complicated problem at hand is to shape microwave signals, which are used to address qubits in a superconducting circuit, in a way to make them do the exact operation on a quantum circuit she is interested in. "You send those signals into the circuit, you change the state of the circuit and that’s how you do a computation," she explains. But not all microwave signals will lead to the desired operations. The computation depends on a variety of variables like the frequency or power of the microwave signals. To try and optimize the shape of the microwave signals, Emily uses machine learning. Specifically, reinforcement learning, a type of machine learning where no model of the system is needed. Because quantum hardware is complex and still error prone, it is hard to simulate exactly, as the Ph.D. student explains. "I chose reinforcement learning, because it’s a program that can learn directly on the hardware and take those errors into account."
The optimization Emily is aiming for, is concerned with the speed of the quantum gates. "One thing we want the gates to do, is to get them to be shorter – so that we can do more gates in less time," she explains. Quantum information has a finite, short lifetime before decoherence happens. “Eventually, the energy leaks out of your system and you don’t really have that information anymore. That’s why you want to do as many operations as possible before that happens." But designing faster gates comes with another challenge: "To have a faster gate, you need to inject more energy into your system, which tends to cause errors," Emily adds. That is where her machine learning approach comes in – to try and find a way to optimize these two competing effects. The Ph.D. student is optimistic that optimized, fast quantum gates could be easily half the time of the current industry standard. And how fast are 'fast' quantum gates? Well – very fast. "We are talking on the nanosecond scale," Emily explains, "a hundred million times faster than you blink, approximately."
Designing faster quantum gates with reinforcement learning was already the topic of Emily’s Master’s thesis. During that time, she worked on the algorithm and the theory behind it. For a proof of concept, she was also able to test it on IBM’s superconducting hardware. "I was able to show once, that my algorithm did work and made fast gates that had a good quality," the Ph.D. student recounts. "However, it was not enough to claim that it is something that everybody should now be using to make their gates." Which brings her to something she loves about doing her Ph.D. at the WMI – everyday access to the labs and quantum hardware. Her first goal is to prove more definitively that her algorithm is useful and successful for single-qubit gates. Later down the road, she hopes, her machine learning program will also be able to work on a two-qubit gate.
Emily Wright, 24
Position
Ph.D. student
Institute
Walther Meißner Institute (BAdW)
SQQC
Degree
Applied Mathemathics & Physics
Emily is working on the development of fast quantum gates. She uses machine learning to optimize microwave signals so that they perform exactly the desired operation in a quantum circuit.
From theory to reality
Solving exciting engineering problems, as Emily puts it, is one of the main motivators for the 24-year-old. When she first got into quantum computing, she found it interesting that everything can be defined with "relatively simple linear algebra" – her background in Mathematics clearly showing. "To go from this theory, that is really contained and easy to work with, to the reality of such a huge engineering problem is what keeps me going", she says. "I mean, how can we get what is on paper working in our real life, in our lab."
To answer this question, she also collaborates closely with Steffen Glaser’s theory group. They are also working on control problems to make better – in this case that means more robust – quantum gates, using a numerical approach. Quantum gates have to be robust against fluctuations in the parameters of the quantum chip, for example fluctuations in the frequency of the qubit, to still be able to do the desired operation even when the microwave signal is slightly off-frequency. "They work on how to make our microwave signals robust against errors," Emily explains. "However, the reality is, these numerical simulations are not entirely able to capture the dynamics of a real quantum processor." Emily is testing the pulses she gets from Glaser’s group experimentally: "I am trying to calibrate them to get them working on our quantum chips, then I am thinking about how we can improve the simulations to get closer to the reality." The aim for the future is to be able to just numerically optimize a pulse and immediately have it working on a quantum chip.
Knowing both sides of the same problem – the theoretical side as well as the experimental side – Emily finds it hard to decide, which one she prefers. "I think I’ve been trying to answer that question my entire career and haven’t figured it out yet," she says laughing. And maybe it is not necessary to make a choice. All throughout her academic life she has been on projects that had both elements. And both sides have certainly advantages and disadvantages. "There is the frustration of your equipment breaking but there is also the satisfaction of seeing something concrete happen. And in theory there is the fun of getting to let your mind wander and explore the problem. But there is also the challenge of not having a solution, not even having proof that you are going in the right direction. So, I think it is good to be able to trade them off and also to see them work together."
A typical day for Emily consists of solving equations and writing code – and then going one floor down from her office to the lab and running her experiments. Besides that, the maintenance of the cryostats to cool the superconducting hardware down to a few millikelvin – "so, really, really, really cold" – takes up a lot of her time: "In fact, this afternoon, I think I will spend several hours working on maintaining one of the pumps that we need for our fridge’s cooling system."
Adventures in and outside of academia
The biggest challenge for the Ph.D. student so far is keeping up with the literature and research of every other group and company that is working on similar problems. "It is such a booming field right now," she explains, "and there are hundreds of publications every week that I need to look through because I want to make sure that my research is still new and relevant."
But, expanding her knowledge every day is also something Emily enjoys a lot. That she had the opportunity to broaden her horizon and was able to switch fields from Mathematics into Physics makes her particularly proud. "I walked into my Master’s in Physics, having not taken anything beyond classical mechanics in Physics. And suddenly I am in a Master’s level quantum mechanics course," she recounts. "Luckily, it is all just Maths underneath," she says laughing. That you absolutely are able to change your topic and expand your horizons by learning new things is a message, that is really important to the Ph.D. student.
It also hints at her adventurous side. First moving across Canada from the East to the West for her Master’s, then moving to Munich for her Ph.D. – which, in terms of flight time, is actually not that much further away from her home town – speaks also a great deal of that. Adjusting to her life in Germany, Emily found pretty easy. "My colleagues, many of them are now friends, are lovely people. I had a great welcome here," she tells. Leaving her family behind in Canada had been of course challenging, especially her twin sister, whom she is really close with. "But, overall, it just has been too exciting for me to miss anything really from Canada," she says with a big smile.
The proximity of Munich to the mountains certainly helps, as a good part of Emily’s hobbies takes place there: rock climbing and hiking, ski touring and skiing is what she loves to do in her free time. And at some point, she hopes, she can bring her mountain bike from Canada. "I mean, Canada is sort of famous for its nature and wilderness and so it is still important to me to have that kind of access", she says. But at the moment, her favorite pastime has nothing to do with the mountains: "Right now, I am mostly playing frisbee, which is super fun!" She joined a team and plays tournaments on the weekends.
Building something magical
Not only in sports Emily appreciates the community she has in Munich and the quantum community that Munich Quantum Valley offers is no exception from that. "I always get the feeling, that if I had something I was unsure about or something I wanted to collaborate on, that there is no doubt I could find somebody to do it with," the Ph.D. student says. And it is not only the opportunity to collaborate with other institutes, but especially the joint expertise within the Walther Meißner Institute that amazes her on a daily basis: "I love the fact, that we build our own chips. And then put them in our own fridges. And then do our own research on them. And that in order to do that, we have so many experts in different things here." To her it is important to emphasize, that building quantum computers is a huge team effort: "Building a quantum computer is not a solo task. And I think that it is a great privilege that I get to work here with so many people who come together to somehow build something that I still think is a little bit magic," she continues.
Looking further in the future, after her Ph.D., Emily can see herself working in industry, despite loving academia very much. Teaching is what she particularly enjoys, she hasn’t lost that as she’s grown older, as she remarks: "The best way to make sure that I know the material is having to tell other people about it. And also, hearing interesting questions from the students, trying to figure out how to answer them." But, working as an engineer after she graduates is what she states as her priority. "I like the very defined projects and goals that you can find in industry, that are sometimes missing in academia," she adds. Moving back to Canada, closer to her family again, is at the moment a likely option for the Ph.D. student. Canada has a good quantum ecosystem – but so does Germany, she ponders: "I really have no idea right now, but I think there is no wrong answer to it. It’s gonna be great either way." And with her positivity and enthusiasm, it certainly will be.
Published 27 September 2024; Interview 02 July 2024