April 1, 2019

Aiming for practical application of quantum annealing machines

System Platform Research Laboratories
Tomohiro Yamaji

Tomohiro Yamaji majored in physics at university and earned a doctoral degree in experimental subatomic physics. Since entering NEC in 2018, he has been involved in the research and development of quantum annealing machines. He is currently engaged in superconducting circuit design, measurement and analysis, with the aim of developing practical applications by 2023.

Aiming for practical application of quantum annealing machines by 2023

I am presently working on the R&D of elements used in quantum computers at the System Platform Research Laboratories in Tsukuba. There are two types of quantum computers. One type uses the quantum gate method, which has been thoroughly researched for a long time, and the other type uses the quantum annealing method, which has been attracting attention lately. Our team is currently focused on computers that use the quantum annealing method.
I am in charge of the hardware used in these machines, and I am performing design, theoretical calculation, measurement, and analysis of circuits that use superconductors. NEC is currently aiming to develop practical applications for quantum annealing machines that use these superconducting circuits by the year 2023, and is now expanding the organization to achieve this aim.
Quantum computer technology has a fairly close affinity with AI technology, which requires extremely high volumes of computation. However, computers that use conventional methods are soon expected to reach a peak in terms of their computation processing capabilities. In this respect, the extremely high computation capabilities of quantum computers can be leveraged for situations such as combinatorial optimization problems that are required in AI. For this reason, quantum computers have attracted a great deal of attention, and research institutions and companies around the world have been concentrating efforts to develop them since the 2000s.
If high-speed processing takes effect for combinatorial optimization problems, then shipping companies, for example, can quickly obtain solutions to problems such as finding the best order to load trucks, or figuring out the most efficient way to deliver packages, which can lead to the achievement of great streamlining in the logistics field. Of course, this is just one example. If quantum annealing machines are completed, I believe it will create an enormous impact by further advancing the spread of machine learning and artificial intelligence in society.

Prolonging quantum effects and enabling scalable design

As a company, NEC originally took the lead in the research and development of quantum gate computers. The assets accumulated through this research have been very useful in quantum annealing research. In addition to research expertise, the strong network in academia, which even includes the famous researcher who created the first quantum bits in the quantum gate method, is a great asset, too. Also, the ability to collaborate with institutions such as AIST (National Institute of Advanced Industrial Science and Technology) and RIKEN (comprehensive research institution in Japan) to perform cutting-edge experiments and theoretical calculation, while also receiving support from METI (Ministry of Economy, Trade and Industry) and NEDO (New Energy and Industrial Technology Development Organization), is a powerful force in propelling our research forward.
Also, in machines that use NEC's unique superconducting circuits that I am in charge of, one characteristic is that the quantum effects (coherent states) have a long duration. In the past, the short duration of the quantum effects was a technological barrier faced by the quantum annealing method, even compared to the quantum gate method. As a result, the quantum effects often did not persist until the computations were complete, thereby limiting the ability to sufficiently utilize the advantages as a quantum computer. In this respect, the quantum state can be sustained longer in the quantum annealing machines that use our superconducting circuits, which is a promising sign in terms of the ability to appropriately use the quantum effects required for computation.
Another significant characteristic is that our machines can be designed in a scalable manner. By adopting the idea of making physical quantum bits redundant and establishing connections between all physical quantum bits, the problem of wiring between quantum bits was also eliminated, making it possible to apply the technology on a large scale.
Together as a team, we are making a concerted effort to establish our quantum computer as the new standard. I find great motivation in working toward the completion of an innovative machine that can change the world.

Research environment that supports freedom of inquiry

I studied physics up to the Ph. D. level, and was engaged in experimental subatomic physics. In subatomic physics, the theories are extremely clear, which is something that fascinated me about the research. I like it very much that you can take a theory and conduct experiments to demonstrate its validity. In terms of the research field itself, my current research is probably in a slightly different area. However, superconducting circuits are created based on calculations and theory, so there are natural similarities. I am enjoying the research very much.
What prompted me to choose NEC is that I am interested in its wide range of business areas. The company is involved in everything from undersea cables to satellites, so I thought that I would be able to do something interesting. Also, at my interview, I was asked if I would be willing and able to work on things like quantum computer simulations. I thought it sounded like fun, and I told the interviewer that I could do it. That is ultimately where I ended up.
About one year has passed since I entered NEC, and I feel that there is a lot of freedom in the environment, which makes it easy to perform research. There is a rule that you can devote about 15% of your research to themes that fit your own personal interests, which gives you some room to do what you like. I like to play around with Python to relax, and recently I got together with the software research team to think about combinatorial optimization problems. Although I am in charge of hardware, I like to tinker with programming.
Another recent aim of mine is to use up my paid vacation time. I was too focused on research in the previous period, and I found that I had accumulated a lot of vacation time that I needed to take. I believe in the importance of taking a rest to improve the efficiency of my work, so I want to make an active effort to take my holidays.