Working in Quantum. Insights and behind-the-scenes with ParityQC’s Matthias Traube

Working in Quantum. Insights and behind-the-scenes with ParityQC’s Matthias Traube

Welcome back to our Working in Quantum interview series, where we put the spotlight on the people that are part of the flourishing quantum computing industry! In this new interview we talk with Matthias Traube, who is co-leading the Use Case department here at ParityQC.

Matthias has a substantial academic experience at the intersection between maths and physics. After earning his Master’s and PhD from LMU in Munich, he worked as a researcher at the Max Planck Institute for Physics and at the University of Hamburg. With quantum computing, he’s found a field in which his knowledge perfectly comes to fruition.

What do you do at ParityQC?

At ParityQC I’m doing mostly two things: I’m co-leading the Use Case Department and besides that I’m heavily involved in writing the compiler for the QSea quantum computers in Hamburg. As the name suggests, the Use Case Department is all about use cases for NISQ hardware. We are investigating which quantum algorithms work better, why they work (or don’t work) and how the type of use case influences the quantum algorithm choice.

The work in the QSea project is somewhat different. Similar to classical computing, when trying to run an abstract quantum circuit on a given QPU one has to translate the quantum circuit to native machine instructions. There isn’t a unique sequence of machine instructions for a quantum circuit, but many different possible ones. The tricky part in compiler development is to come up with strategies to produce optimal machine instructions, and the QSea ion-trap quantum computer has quite special native interactions, making the task even more exciting.

Tell us about the path that led you to working in quantum computing.

My career path is a circle: I started to study physics, started to branch off to mathematics, did a PhD at the interface of math and physics, followed by a PostDoc in math and now I’m back at physics. My academic research was in conformal/ topological field theories in two and three dimensions. Besides beautiful math, they also have a connection to condensed matter physics and actually to topological quantum computing or topological quantum error correcting codes. Hence, when I hit the abort button for my academic career, I thought quantum computing is actually as close as it gets to applying my math background in “real life” and here I am.

Among the topics you studied and researched, besides the foundations of theoretical physics and mathematics, which specific topics turned out most useful in your current work?

I did nothing else than studying the foundations of theoretical physics (string theory) and mathematics. But jokes aside, I did a lot of research on so called string-net models, which are a way to explicitly compute certain objects in topological field theories. They originated in physics and a special case of string-nets can, in principle, be even used to get a topological error correction code. It is more on a meta-level, though, my research background helps for my current work. E.g. I did a lot of graph manipulations on higher genus surfaces, which now becomes quite handy to understand graph theory concepts which are ubiquitous in quantum computing.

What are the main goals of the Use Case department?

The first and foremost goal is to get our ASQ tool running smoothly. ASQ is an amazing modular framework for defining and comparing NISQ-era quantum algorithms. All other goals derive from that. I can give two examples:

  • Whenever you have a combinatorial optimization problem and you want to know whether a NISQ-era algorithm may work for it, you can simply put it in our ASQ tool and it will give you the best quantum algorithms for it, together with a performance estimate.
  • On the other hand, you may have developed a new NISQ algorithm and you want to know how it compares to existing ones. Then you can implement it in the ASQ tool and run the comparison to see for which type of problem the new algorithm may offer an advantage.

Recently, there have been a number of breakthroughs in research that have shortened the timeframe for reaching useful quantum computing. How has this influenced the field of quantum use cases? Is it possible to identify more clearly in which industries quantum computing will be most valuable?

Honestly, this is hard to tell and exactly one of the questions we try to address with our ASQ tool. By comparing a vast selection of industry-relevant problems we try to identify patterns of use cases and the corresponding quantum algorithms which offer an advantage.

Last year, you worked on the paper “Encoding-independent optimization problem formulation for quantum computing”, presenting a novel framework for formulating optimization problems for quantum computing. How did this research come together and what are the next steps?

The basic question was how to automate quantum algorithm selection. For this we needed some abstraction and common building blocks for optimization problems. Of course, there was already literature in that direction, but we described a holistic framework which is suitable for automatically generating a quantum algorithm form the very top, the mathematical formulation, down to the very bottom, like QPU specific features. This evolved in the ASQ tool and the next steps are to get that running smoothly.

Of all the potential use cases for quantum computing, which one are you most excited about?

Personally I’m most excited about hamiltonian simulations and use cases in quantum chemistry. If we could improve e.g. drug discovery processes with quantum computing, that would be absolutely amazing. But I don’t want to oversell here. It is still a long way towards anything in that direction.

How do trained researchers like you manage to combine job responsibilities with the need to always learn something new and stay up-to-date on the breakthroughs in the industry?

I have the habit of checking the relevant arXiv sections in the morning. In mathematics this was easy, as there weren’t so many articles. The quantum computing section can be long, though. However, it suffices to go through the titles and maybe read some abstracts to catch something interesting. Besides that, I can be certain that somebody in the company will post interesting papers in our internal chat. It is almost like a company-swarm-intelligence web crawler for interesting papers.

What do you like the most about life in Hamburg, and what fills your days outside of work?

Hamburg is truly an amazing city. You get to see all kinds of people and with all the water and parks it offers nice contrasts. Outside of work, I’m mostly training, either on my road bike or in the gym. I love to go on long bike rides on the weekend, it is almost a meditative state when I’m on the bike. Other than that, I’m a sci-fi literature nerd, especially space-operas. The more exotic alien cultures and faster-than-light travel or wormholes, the better.

What inspires you to come to the ParityQC office (or log in) every day?

The most important thing is, of course, that it is a whole lot of fun. As you already mentioned, quantum computing is rapidly evolving and there is the possibility to learn new things almost daily.? Quantum computing is in a state classical computing was in the 1940s or 1950s. Making a tiny contribution to the foundations of it is just cool.

Can you give 3 pieces of advice for someone just starting out in this industry?

The two standard pieces of advice for everybody in life also hold for someone starting out in the industry: be curious and motivated. Quantum computing is such a new technology, that one doesn’t need be a die-hard expert to work on it (I wasn’t for sure). If you have physics, math or computer science skills, there will be certainly problems you can work on. Third piece of advice is of course to start at ParityQC.

要查看或添加评论,请登录

社区洞察

其他会员也浏览了