Powering the next generation of discovery Quantum Simulator Challenge
Technology News | 2024-03-19
8 minute read
Introduction
The computational power of quantum computing has the potential to radically transform many fields and industries from material science to pharmaceuticals and finance bringing benefits to both enterprises and society.
Unlike classical computing, which relies on bits that can only be in one of two states (0 or 1), quantum computing uses quantum bits (qubits) that can exist in multiple states simultaneously. This allows quantum computers to perform a range of calculations far faster than classical computers, making them ideal for solving many complex problems that may be beyond the capabilities of traditional computing systems today.
The challenge today is that truly practical quantum computers are still some years away from becoming a reality as a range of significant technical challenges remain to be overcome. The complexity of today’s development quantum systems also makes them challenging to operate.
However, organizations can already take advantage of the problem solving capabilities of quantum based computing today, thanks to the availability of quantum simulations.
Quantum Simulation
Quantum simulation can offer organization many of the same potential advantages and benefit as practical quantum computers will once they become available. Quantum simulations use the same quantum principles to enhance the performance and efficiency of traditional binary computing systems in solving complex problems.
Quantum simulators are already enabling leading organization to start taking advantage of quantum methods and principles in solving complex problems. Fujitsu is not only one of the leaders in the race to develop practical quantum computers but also a leader in quantum inspired computing.
Fujitsu was a pioneer in quantum simulation technology and has already developed a 39-qubit CPU-based state vector quantum simulator system consisting of 512 FX700 nodes, using A64FX processors. The quantum simulator utilizes “Qulacs,” one of the world's fastest quantum simulator software developed by Osaka University and QunaSys Corporation. Fujitsu enhanced Qulacs so that it can be executed in parallel and distributed.
Fujitsu has also developed a way on the cluster system utilizing MPI (Message Passing Interface) technology and optimizing memory bandwidth using SVE(Scalable Vector Extension) operation. Fujitsu has also prepared a software development kit (SDK) partially compatible with Qiskit, one of the major development tools for quantum computer software, to offer quantum software developers a highly convenient development environment.
Gaining a competitive advantage
Quantum simulation is also playing a key role in enabling organizations to benefit from the advantages of quantum based calculations and problem solving today. It is also importantly helping them prepare for the arrival of true quantum computers in the future.
Organizations are able to develop quantum based solutions now that run in emulation on today’s hardware and deliver value. When true quantum computers do become available these organizations will be ready and able to immediately migrate to true quantum computer hardware and take full advantage of the extra power and performance these systems will provide.
It can take considerable time for organizations to develop quantum solution and so developing solution using quantum emulators will give organizations a considerable competitive advantage over organizations who wait for quantum computers to become available before starting their own development.
Fujitsu is actively engaged in developing the practical quantum computers of the future, as well as quantum simulations for our present-day High-Performance Computing (HPC) systems. This dual focus provides Fujitsu with a unique perspective on how to assist customers in maximizing the benefits of quantum adoption, both in the present and the future.
The Fujitsu Quantum Simulator Challenge
To help stimulate quantum simulation innovation in February 2023 the Fujitsu Business Incubation Division, a part of the Fujitsu Technology Strategy Unit launched the first Fujitsu Quantum Simulator Challenge for industry and academia.
The Fujitsu Quantum Simulator Challenge offered a prize fund worth $100,000 USD as well as the opportunity to partner with Fujitsu to bring innovations to market. Further investment opportunities both from Fujitsu and the Fujitsu VC network.
It also came with a range of other benefits such as support and training form leading Fujitsu Quantum experts and an opportunity to be showcased at the next Fujitsu ActivateNow Technology Summit.
Participants
The Quantum Simulator Challenge attracted a wide range of international participants form across academia and enterprise who wished to see how quantum simulation could be used to address a wide range of different challenges including:
- Credit card fraud detection
- Quantum algorithms for fluid dynamics
- Quantum Fuzzy Inference Engine (QFIE) for smart city problems
- Distributed energy resources optimization problem
- Hyper-Personalized portfolio optimization
- Multiphysics engineering through Partial Differential Equations (PDE)
- Alzheimer’s disease diagnosis and management
- Antimalarial drug discovery
- Histopathological cancer detection
- Multi-split-step quantum walk (MSQW) on financial market simulation
- Quantum Algorithm for Lattice Boltzmann (QALB)
- Macroscopic Quantum Tunneling (MQT) for novel many-body physics
- Quantum robust fitting (computer vision and pattern recognition)
- Renyi Entropy on large systems with randomized measurement
- Experiments in Quantum memory and stability
- Trial access to MPI Qulacs Simulator
The winning participants
The first Fujitsu Quantum Simulator Challenge effectively demonstrated the advantages and possibilities that quantum simulation can already offer to a wide variety of industries and organizations today.
Fujitsu was thoroughly impressed by the exceptional quality of quantum simulation work presented by all participants, making the selection of an overall winner and runners-up a challenging task.
By harnessing the capabilities of the Fujitsu Simulator, the participants were able to show significant speed up in run-time and performance in their simulations, providing many interesting insights and revealing the potential and possibilities of quantum computing.
After meticulous deliberation, the team from Quanscient OY of Finland was declared the winner. Riverlane, with teams from Cambridge, UK, and Cambridge, MA, USA, secured a close second place. The teams from Qkrishi and the University of Naples Federico II shared the third place honor.
We extend our heartfelt gratitude to all participants for their remarkable contributions and active participation. We eagerly anticipate future collaborations and continued engagement with you all.
Participants research overview
Quanscient OY, (Finland)
First place - Fujitsu Quantum Simulator Challenge
Fluid dynamics the study of liquids, gases, and plasmas in motion is a complex field its understanding is critical in a wide range of fields including engineering, medicine, energy, and aerospace. Quanscient investigated how Quantum simulations could be used in the study of Fluid dynamics to better model and predict the motion of fluids.
The study implemented 1D and 2D quantum algorithms for advection diffusion and Navier Stokes equations on a simulator backend. The execution time and the increase in the number of gates with the lattice size were analyzed. The simulator backends were efficient, allowing for the simulation of one timestep of the 1D ADE algorithm for about 33 qubits and the 2D ADE algorithm for 28 qubits. This provided valuable insights into the algorithms. The results, carried out on the qiskit-aer backend and compared to the analytical solution, showed some discrepancies in the first timestep. However, these were not solely due to the quantum algorithm but also due to the low resolution of the computational mesh, a common issue when using numerical methods to solve differential equations.
The quantum lattice Boltzmann algorithm for one-dimensional advection-diffusion equation and Navier-stokes equation is one step in the direction of simulation of complex fluid phenomena using the revolutionary technology of quantum computing. The significance of testing these quantum algorithms has grown in importance due to the increasing diversity in the availability of quantum hardware and the need for effective algorithm implementations. This study showcases the speed and efficiency of the Fujitsu simulator to test the quantum algorithms for more than 30 qubits.
Riverlane (UK & USA)
Second place - Fujitsu Quantum Simulator Challenge
One of the biggest challenges in developing quantum computers is that quantum bits (qubits) can easily lose their quantum state which makes it challenging to maintain the stability of quantum computations.
This makes developing more advanced quantum error correction key to developing larger practical fault tolerant quantum computers. Riverlane uses quantum simulation to help develop improved quantum error correction techniques.
Leakage out of the qubit computational subspace is a threatening source of noise for quantum error correction. The project used Fujitsu’s state-vector simulator to run simulations of QEC stability experiments under a coherent noise model with leakage, simulations of such detailed error models in stability experiments are novel. They proposed several l “leakage reduction units” to limit the damage of leakage out of the qubit computational subspace. They also developed a way to simulate qutrits within mpiQulacs, extending the capability of the Fujitsu simulator to higher-dimensional systems.
Qkrishi (India)
Third place - Fujitsu Quantum Simulator Challenge
The Financial sector is anticipated to be one of the key markets for quantum computer adoption. Qkrishi looked at applying quantum simulations in the financial sector for financial fraud detection with a focus on Credit card fraud detection.
Qkrishi has been able to develop Sequre a software application that utilizes the Fujitsu Simulator powered quantum machine learning model for financial fraud detection. The backend is built over the Flask library and is hosted on Amazon Web Services, which communicates with the Fujitsu server over a secure layer. It can trigger the model training action on the Fujitsu server, aggregate the model classification results back, and feed the software user interface. The user interface has state-of-the-art react elements and an extensive dashboard. The user can visualize the fraud detection results via this dashboard. It also provides a feature to trigger the model training job and auto publish the results on the dashboard. The front-end dashboard comprises data visualization features such as filtering the results based on a column value.
Qkrishi tested many models, including Quantum Support Vector Classifier(QSVC), QSVC with Quantum Kernel Trainer(QKT), Auto QSVC, Ensemble of QSVC and SVC, and Classical SVC. They have identified the best models, the Auto QSVC model outperforms all in identifying fraudulent activities with high accuracy and recall. As a result of its remarkable performance, it is the preferred choice for robust fraud detection due to its reliability and effectiveness. Sequre represents a significant advancement in the fraud detection domain, driven by its exceptional performance and intuitive user interface. By harnessing the capabilities of the Optimized Auto QSVC algorithm and Fujitsu Simulator, they have surpassed the capabilities of IBM quantum hardware, showing significant speed up in run-time and performance.
University of Naples Federico II (Italy)
Third place - Fujitsu Quantum Simulator Challenge
The increasing urbanization of populations and the need for more sustainable living are driving the development of Smart Cities to improve the management of key infrastructure and services such as transportation, energy, waste management, and public safety. However, the sensors and systems needed to create Smart cities also create vast amounts of data that need to be securely processed often in the near real time to support decision making to keep smart cities running smoothly.
The University of Naples Federico II investigated using a Quantum Fuzzy Inference Engine (QFIE) based on fuzzy logic running on a Quantum simulator to enhance data analysis and decision-making for smart cities. QFIE has been applied to three use cases in the context of smart cities: ALE in sensor node localization process in WSNs, energy efficiency forecasting, and traffic junctions control system. In all the use cases, QFIE achieves higher performance than classical approaches.
This new quantum algorithm, QFIE, recently introduced by the team, enables the execution of interpretable fuzzy systems on quantum computers and provides an exponential computational advantage in assessing fuzzy rules when compared to the standard algorithm used on classical computers. Behind the computational advantage, QFIE also allows for the programming of quantum computers with words, leading to a higher degree of simplification in quantum computation.
Who should be looking at quantum simulation
The organizations who stand to benefit most from early quantum adoption include:
- Pharmaceuticals and Biotechnology
- Finance and Investment
- Cryptography and Cybersecurity
- Materials Science and Chemistry
- Supply Chain and Logistics
- Modeling and Simulation
- Aerospace and Defense
- Telecommunications
- Transportation
- Artificial Intelligence and Machine Learning
These are also the organization at the most risk of comparative disruption if they wait for practical quantum computers to become available before stating their development of quantum based solutions.
How can quantum benefit organizations
Quantum computers will not directly replace today’s classical computers. Organizations will continue to operate classical computer systems for day to day computing for the foreseeable future. However, quantum systems will increasingly be used by organizations who need the problem solving computational power of quantum systems. Quantum computers can benefit organizations in five key ways:
1. Advanced data analysis
Quantum computing can offer faster data analysis than classical computers, enabling real-time processing of large datasets for businesses. This leads to better decision-making and market adaptation. For example, it can reveal customer trends, helping businesses tailor products and services to meet demands more effectively.
2. Quantum enhanced discovery
Accelerating the discovery process and reducing costs, revolutionizing a wide range of scientific fields, such as modern drug discovery, which involves analyzing vast amounts of complex data, a complex and time consuming process.
3. Quantum enhanced forecasting and simulations
Quantum computing's speed and power can be used to significantly improve complex forecasting and simulation models, such as climate change and weather predictions, by analyzing vast data from multiple sources. In financial markets, it offers a competitive advantage through faster, deeper insights and more accurate predictions based on numerous variables.
4. Quantum optimization
Quantum can help us develop more efficient and sustainable products and businesses by improving production design, reducing resource utilization and optimizing supply chain logistics.
5. Quantum encryption
Using quantum mechanics properties like entanglement and superposition we can develop virtually unbreakable Quantum based encryption transforming the cybersecurity landscape and enabling businesses and governments to safeguard sensitive information and critical infrastructure from advanced threats.
Conclusion
Quantum simulation allows organizations to access many of the benefits of quantum today. It also allows organizations to prepare to take full advantage of quantum hardware once it arrives and to gain competitive advantages over organizations that delay development until quantum hardware is available.
Fujitsu is one of the leaders in quantum simulation, quantum inspired computing and quantum computing. Talk to us and find out how we can help your organization take full advantage of quantum.
Key recommendations
- Organizations who are involved in high-performance computing fields such as financial, scientific and industrial research and analysis or national security and defense should be looking now at how that could use quantum inspired computing to gain and maintain new competitive advantages and avoid potential future disruption driven by late quantum computing adoption.
- Discover more about Fujitsu’s research into quantum computing at: https://www.fujitsu.com/global/about/research/technology/quantum/
- Find out more about Fujitsu quantum inspired computing at: https://www.fujitsu.com/global/services/business-services/digital-annealer/services/
Why not talk to Fujitsu and see how we can help your organization?
Nick Cowell
Principal Consultant/ Technology Strategy Unit, Fujitsu
Nick is a technologist with extensive experience in hardware, software and service development, having previously worked for leading technology providers across the USA, Europe and Oceania.
