Results of the FY2023 MITOU Target Program: Software Development Utilizing Quantum Computing Technology

In-page Links

Results of the FY2023 MITOU Target Program: Software Development Utilizing Quantum Computing Technology

Category 1: Software Development for Annealing Machines

PM: TANAKA Shu

Art museum Tour Route Optimization Application Using Annealing Machines

We developed an art museum tour route optimization application using an annealing machine. This application utilizes quantum computing technology to optimize art museum routing. The goal of the project is to provide exposure to quantum computing technology and to increase the number of people interested in this field. It also aims to contribute to applied research in Quantum Annealing. The application is available for smartphones and presents the best route for the user's setup such as tour time and walking speed. The utilization of optimized routes has been evaluated to increase satisfaction. The source code for this application is shared on Github and is freely available to all.

• View the details of Art museum Tour Route Optimization Application Using Annealing Machines (in Japanese)

Creators: KOMIYAMA Tomoko, OGAWA Yudai

Development of an automated shift allocation service for tutoring schools

This project aims to develop an automated shift scheduling service for individual tutoring schools. Tutoring schools are operated in two forms: group tutoring schools and individual tutoring schools. In individual tutoring schools, the learning content can be provided to each student by own teacher who has limited the number of students. In individual tutoring schools, it is necessary to create shifts for instructors and students during the regular course period, which is a burdensome task for tutoring schools.
Existing shift grouping software performs manual and semi-automatic assignment, but it is difficult to search for the best combination, and the full search is time-consuming. The project aims to shorten the working time and improve the accuracy of shift-assignment by developing an automated shift-assignment web application called "Zyuku-Koma" for individual tutoring schools, as well as to verify automatic shift-assignment using quantum annealing.

• View the details of Development of an automated shift allocation service for tutoring schools (in Japanese)

Creator: FUKUDA Shuichi

PM: TANAHASHI Kotaro

Development of a tool to support exploration and analysis of alloy catalysts using an annealing machine

This project aims to develop a tool to support the metal configuration search and analysis of alloy catalysts using an annealing machine. Alloy catalysts are important for achieving carbon neutrality, and optimal metal configurations must be explored. This project has two objectives: first, to explore metal configurations of binary and multi-alloys using an annealing machine to find the optimal configuration; and second, to analyze and visualize the explored metal configurations to obtain information needed by catalytic chemistry researchers. The developed software allows the optimization of metal configurations using an annealing machine and visualization of formation energies and physico-chemical analysis results. This tool will be useful for catalyst researchers and is expected to accelerate the development of alloy catalysts. In addition, the results of this project are expected to be applied not only to catalyst research but also to the creation of highly functional materials, which will contribute to the conversion of carbon dioxide into valuable chemicals, hydrogen into energy, and the development of new multi-element materials.

• View the details of Development of a tool to support exploration and analysis of alloy catalysts using an annealing machine (in Japanese)

Creators: SANPEI Hiroshi, SAEGUSA Koki

Development of ZEB Design Support Tool for Non-Residential Buildings

This project aims to develop ZEBOpt, a Net Zero Energy Building (ZEB) design support tool for non-residential buildings. ZEBs are buildings that aim to achieve a zero energy consumption balance, which is important for achieving carbon neutrality. However, evaluating ZEBs requires manual compilation of complex building specifications, a time-consuming and labor-intensive task with respect to optimizing energy savings and meeting standards. This project will develop a tool to assist in the design of ZEBs by optimizing the combination of building materials by working backwards from the building energy efficiency index. Quantum annealing technology will be used to efficiently search for the optimal solution and achieve a balance between exterior design and energy load. This is expected to reduce the effort required to evaluate ZEB designs, optimize energy savings, and reduce costs at the same time. The development of this tool is expected to make it easier for designers of non-residential buildings to realize ZEBs and contribute to labor and energy savings in the construction industry.

• View the details of Development of ZEB Design Support Tool for Non-Residential Buildings (in Japanese)

Creators: TOYODA Shoichiro, TOYODA Suzu

PM: TAMURA Ryo

The Development of a School Lunch Recommendation System Using an Annealing Machine

This project aims to develop a school lunch recommendation system using an annealing machine. School nutritionists and dietitians spend a significant amount of time and effort creating meal plans. In this project, we developed a web application that recommends optimal meal plans using an annealing machine. Users can set constraints such as target nutritional values, calories, and costs, and the application solves the combinatorial optimization problem to suggest the best meal plans. Additionally, users can edit the recommended meal plans or add their own original recipes. This application is expected to reduce the burden on nutritionists and dietitians in creating meal plans and to facilitate more efficient meal planning. By leveraging the annealing machine’s ability to quickly find optimal solutions, the application enables more efficient meal plan creation.

• View the details of The Development of a School Lunch Recommendation System Using an Annealing Machine (in Japanese)

Creators: INOUE Yuka, AKISHIMA Haruka

Development of a New Delivery Optimization Method using Annealing Machines

We have developed LRightAway, a novel delivery optimization method utilizing annealing machines. This algorithm optimizes delivery routes by considering consumer waiting times, addressing the increasing demand for deliveries and the challenges of delivery optimization in the logistics industry. The method leverages annealing machines to facilitate innovative delivery formats within local delivery networks between consumers and deliverers. Specific advancements include the integration of information from online supermarkets and order data, followed by optimization calculations using an annealing machine to determine the allocation of deliverers and the sequence in which products are delivered. This method offers significant value to online supermarkets and fresh food delivery services by helping “light a way right away” for efficient and timely deliveries. Furthermore, it has the potential to be applied to drone delivery and automated delivery systems in the future, contributing to next-generation delivery optimization.

• View the details of Development of a New Delivery Optimization Method using Annealing Machines (in Japanese)

Creators: TERASHIMA Yuto, ISHIAI Satoki, HONDA Rio, AOKI Shiori

Category 2: Software Development for Gate-based Quantum Computers

PM: TOKUNAGA Yuki

Quantum education through the development of an automatic quantum program scoring system and Creation of a huge data set of quantum programs

We have developed Qookbook, an automated grading system for quantum programs. This system is a web service for efficient learning of quantum computers, providing learning through frequent exercises and automatic grading. Users can write programs as they read through the text and check their understanding of the concepts through automatic grading. Qookbook also provides a database of quantum programs that can be used for educational and software engineering analysis. Furthermore, Qookbook can also host contests to provide a practical attempt at quantum programming. We hope that the use of Qookbook will foster human resources with knowledge of quantum computing and contribute to the research and development of quantum computing systems.

• View the details of Quantum education through the development of an automatic quantum program scoring system and Creation of a huge data set of quantum programs (in Japanese)

Creator: AOYAMA Koki

Development of Efficient Quantum Machine Learning Algorithms Using Dynamic Circuits

We have developed efficient quantum machine learning algorithms using "dynamic circuits." Quantum computation is currently difficult to put into practical use due to hardware constraints such as noise and small number of qubits, but it is known that these constraints can be reduced by using dynamic circuits. In this project, we aim to apply dynamic circuits to quantum machine learning algorithms to improve existing methods and develop new methods. Specifically, we proposed a method to reduce the number of qubits by using dynamic circuits in quantum convolutional neural networks (QCNN), and also developed a new quantum state generation model with dynamic circuits. These methods are expected to improve the performance of quantum machine learning algorithms and suggest new potential applications.

• View the details of Development of Efficient Quantum Machine Learning Algorithms Using Dynamic Circuits (in Japanese)

Creators: KAZAMA Haruhi, KAMATA Kohei

Development of Interactive Visual Quantum Educational Materials and Quantum Simulation Tools

We are developing an interactive visual quantum educational tool called Qualsimu, along with a quantum simulation tool. This educational material aims at connecting information and physics naturally, and this educational resource starts with the basics of quantum mechanics and quantum information, then moves on to simulations of superconducting qubits and quantum dynamics. The course material makes extensive use of animations and visualizations to help students understand difficult concepts in an intuitive way. The simulation tools also allow students to perform real quantum experiments and compare them with the theory. It is hoped that these teaching materials and tools will enable people interested in the field of quantum information to engage in practical learning and contribute to the development of quantum human resources.

• View the details of Development of Interactive Visual Quantum Educational Materials and Quantum Simulation Tools (in Japanese)

Creators: NAKAMURA Taishi, SUZUKI Taiga, AOKI Koichi

PM: FUJII Keisuke

Development of Boundary Element Method Using Quantum Computers

This project aims to develop the Boundary Element Method (BEM) using quantum computers. BEM is one of the representative numerical analysis methods in the field of computational mechanics and has been applied to various engineering problems. In this project, after acquiring basic knowledge of quantum algorithms, we conducted research on the development of several quantum algorithms necessary for the BEM analysis. We selected the parts of the BEM computational scheme that could be solved using quantum computers and carried out the analysis. There are no examples globally of research utilizing quantum computers for BEM, which demonstrates the high level of novelty of this project. The software developed in this project is intended for researchers and engineers interested in physical simulations, and it has the potential to be used for large-scale simulations in the future.

• View the details of Development of Boundary Element Method Using Quantum Computers (in Japanese)

Creators: SAITO Takahiro

Development of a Quantum Programming Contest System and WebAssembly Implementation of a Quantum Circuit Simulator

This project aims to develop a quantum programming contest platform and convert a quantum circuit simulator to WebAssembly. The contest provides a competitive and educational environment for quantum computing enthusiasts, fostering the growth of new quantum talents. The system is available as a web service, allowing users to learn quantum computing by solving problems. Unlike existing quantum programming contests, this platform is unique in its accessibility to a broader audience. The next goal is to expand the service globally, contributing to the advancement of quantum computing on an international scale.

• View the details of Development of a Quantum Programming Contest System and WebAssembly Implementation of a Quantum Circuit Simulator (in Japanese)

Creator: YUKIYOSHI Kein

PM: YAMAMOTO Naoki

Development of a Hybrid Quantum Error Correction Method Using Model Predictive Control

In this project, we are developing a new quantum error correction method using model predictive control. This enables effective correction of hardware noise in quantum computers. Compared to conventional methods, this method improves stability and responsiveness, and allows correction to be performed while predicting future noise models. Moreover, by combining techniques such as system identification using machine learning, this method has the potential to become a more versatile error correction method. These results are expected to expand the range of applications of quantum computers and enable the design of better quantum computers.

• View the details of Development of a Hybrid Quantum Error Correction Method Using Model Predictive Control (in Japanese)

Creator: KASAHARA Nobuhiro

Development of a Fast Machine Learning Method based on Quantum Reservoir Computing

In this project, we develop a fast machine learning method using quantum reservoir computing (QRC), an emerging computational paradigm that harnesses quantum systems as computational resources. Traditional QRC architectures are hindered by the collapse of the quantum state upon measurement, resulting in the loss of temporal input memories vital for time-series analysis. Our proposed QRC framework is distinguished by the integration of feedback mechanisms based on prior measurement outcomes, effectively restoring the memory of previous inputs even after measurement. We demonstrate the efficacy of our QRC in time-series analysis, and under specific conditions, it outperforms traditional classical reservoir computing models. Our findings highlight the significant potential of QRC in the forthcoming quantum era as a powerful tool for time-series analysis.

• View the details of Development of a Fast Machine Learning Method based on Quantum Reservoir Computing (in Japanese)

Creator: KOBAYASHI Kaito