電気自動車と電池は明日を拓く

切り抜き詳細

発行日時
2020/3/25 8:30
見出し
Fujitsu Laboratories and Quantum Benchmark begin joint research on algorithms with error suppression for quantum computing
リンクURL
https://www.greencarcongress.com/2020/03/20200325-fujitsu.html Fujitsu Laboratories and Quantum Benchmark begin joint research on algorithms with error suppression for quantum computingへの外部リンク
記事詳細
Fujitsu Laboratories Ltd., and Quantum Benchmark Inc. of Canada will conduct joint research on quantum algorithms using Quantum Benchmark’s error suppression technology as they aim to advance the capabilities of current generation quantum computing platforms. Quantum Benchmark, a startup founded by leading researchers from the University of Waterloo’s Institute for Quantum Computing, provides software solutions for error characterization, error suppression, and performance validation for quantum computing hardware. In this collaborative research project, the companies will develop practical quantum algorithms utilizing Fujitsu’s AI algorithm development technology as well as its knowledge gained through Digital Annealer applications in finance, medicine and material development. The Digital Annealer is Fujitsu’s new quantum-inspired architecture that can rapidly resolve combinatorial optimization problems. Overview of the joint research. Quantum Benchmark’s patented True-Q software system, which enables optimal performance of current hardware, is a key to this development. Accordingly, Fujitsu Laboratories and Quantum Benchmark will endeavor to solve problems in the fields of materials science, drug development and finance that are intractable to solve with conventional computers. Quantum computers are expected to be able to perform a new form of computation by harnessing fundamental properties of the quantum world, such as entanglement and superposition. This is often explained by invoking the idea that they can process both 0 and 1 at the same time, and the continuum of states in between 0 and 1. This advantage comes by performing calculations using quantum bits, called "qubits", which is unlike conventional computers which process conventional bits, that can be only 0 or 1. However, quantum bits are fragile and highly vulnerable to errors and noise, and as time goes on, the effects of noise add up, making the quantum calculation results inaccurate. Since calculations for pharmaceuticals and materials are time-consuming, there is a need to develop error-suppression methods enabling algorithms to overcome the effects of noise. Under the partnership, which is slated to run to March 2021, and planned for extension after April 2021, Fujitsu will develop quantum algorithms for applications such as quantum chemistry and machine learning, and develop performance analysis technology for quantum algorithms in simulations. Quantum Benchmark will support the implementation of True-Q error diagnosis technology on current quantum computing platforms; support implementation of quantum algorithms on current quantum computing platforms; and support custom specific error suppression strategies and performance evaluation for quantum algorithms on current quantum computing platforms. Fujitsu Laboratories and Quantum Benchmark will expand the scope of their joint research beyond finance, drug discovery, and materials, as they plan to develop quantum algorithms to be implemented in quantum computers for various applications which could not be solved with conventional computers. The companies aim to demonstrate new applications on a 100+ qubit quantum computer by 2023.