Where Physics Meets Information: Quantum Annealing

The problem of searching for the best option from a vast number of choices while satisfying certain constraints is called a combinatorial optimization problem. Combinatorial optimization problems are all around us. Take, for example, a package delivery service. There is a vast number of possible delivery routes (options). It is necessary to find a route that efficiently delivers all packages while adhering to specified delivery times (constraints). As the number of delivery locations increases, the number of possible routes grows exponentially, making it difficult to find the best route by listing all possible patterns.

Quantum annealing is a computational technique expected to provide better solutions for combinatorial optimization problems. Quantum annealing is an algorithm that represents a combinatorial optimization problem using a mathematical model from statistical mechanics, a field of physics, called the Ising model. It then operates by using quantum effects to search for the lowest energy state of the Ising model. Quantum annealing is truly a computational technique born from the intersection of physics and information. Quantum annealing was proposed in the context of theoretical physics in 1998, and the world's first commercial quantum annealing machine appeared in 2011. Now, a decade later, the field of quantum annealing is taking its first steps toward social implementation, beyond just academic research.

I will leave the details of research and development in the quantum annealing field for another time. Here, I would like to introduce the "encounters" I have had in my research within this field.

I began exploring applications for quantum annealing when I was a graduate student. At that time, I had the opportunity to discuss my research field with a friend from my cohort who was studying machine learning, and we decided to conduct joint research on applying quantum annealing to machine learning. The research itself went smoothly, but at the time, quantum annealing was largely unknown, and machine learning was not as popular a field as it is today, so I struggled with where best to present our findings. Fortunately, our work was accepted at UAI, an international conference on artificial intelligence, in 2009. This brought the research to a conclusion and allowed me to experience both the difficulties and the joys of encountering a different field. Building on this experience, I have recently been fortunate to meet many researchers and engineers, including people from companies in a wide range of industries. I am now engaged in a broad spectrum of research, from foundational studies for hardware and software development in the quantum annealing field to the exploration of new applications.

In April 2020, I was appointed to the Department of Applied Physics and Physico-Informatics, Faculty of Science and Technology, Keio University. Starting in the 2021 academic year, a Project Lecturer, graduate students (including a working professional pursuing a doctorate), and fourth-year undergraduate students have joined my laboratory. Building on the encounters with the members of my laboratory, and cherishing the new encounters to come, I hope to build a new current in the field of quantum annealing together with a diverse range of people.