Scientists at the University of Science and Technology of China say that the Jiuzhang quantum computer they are developing can perform tasks commonly used in artificial intelligence (AI) 180 million times faster than with the most powerful supercomputer in the world today.
Jiuzhang can solve problems in applications of data mining, bioinformatics, network analysis and chemical modeling research. The experimental results were published by the research team in the prestigious journal Physical Review.
Lead researcher Pan Jianwei, a physicist at the University of Science and Technology of China, said that Jiuzhang is used to solve outstanding real-world problems that traditional supercomputers have not been able to solve. reasonable.
In one experiment, the team used Jiuzhang to solve a problem that is difficult for traditional computers. In the process, Jiuzhang used more than 200,000 samples and took less than 1 second to process. Meanwhile, making each model the fastest traditional supercomputer in the world would take 700 seconds, which is almost 5 years to process more than 200,000 samples.
In theory, quantum computers are much faster and more powerful than conventional computers. However, the subatomic particles at the heart of this technology are fragile, short-lived, and even a small disturbance from the surrounding environment can cause errors. So to avoid interruptions, most quantum computers operate in extremely cold and isolated environments.
Unlike other quantum computers, Jiuzhang uses light as the physical medium for calculations, so it does not need to operate at extremely low temperatures in a closed environment and operates stably for longer.