Recently, The Australian quantum company Q-CTRL announced the latest quantum automatic error correction software tool, Fire Opal. The company's benchmark experiments on superconducting quantum computers have shown that Fire Opal software can make quantum algorithms more successful on hardware than the industry's best compiled software to perform quantum algorithms.
Image from Q-CTRL
The development of quantum computing needs to greatly improve the performance of quantum computers, so that quantum computers can be used by enterprises, so as to solve complex problems in finance, materials science and other fields.
But most quantum computers are still prone to error, hindering the widespread application of quantum computing. In this regard, the Q-CTRL team found a new way to greatly increase the success rate of quantum algorithms through self-developed automatic error correction technology, and achieve the maximum performance of quantum computers as much as possible, thereby narrowing the gap between the promised performance of quantum computers on the market and the actual performance they deliver to users.
In the long run, the company said, combining quantum computers, circuit optimization and quantum error correction technologies will be the key to achieving large-scale quantum computing in the future.
Q-CTRL conducted benchmark experiments on Fire Opal software, including the Bernstein-Vazirani algorithm benchmark established by the American Quantum Economy Development Alliance (QED-C). In this algorithm, the Fire Opal software demonstrates computing performance that surpasses other competitors. And as the quantum computer grows larger, the performance improvement is also higher, growing exponentially. When the computer reached 16 qubits in size, the Fire Opal software achieved an algorithmic advantage of 9,000 times that of the competition.
Bernstein-Vazirani algorithm benchmark results, image from Q-CTRL
In the Grover quantum search algorithm ( an unstructured search algorithm that runs on a quantum computer ) , using a computer consisting of 5 qubits as an example , Fire Opal software is able to binary optimize the results of calculations , turning a "completely wrong" solution into the correct one. Although in this case, the algorithm success rate is only increased by about 20 to 30 times, the effect is obvious.
Paulina Mazurek, CEO of BeIT, an early customer of Q-CTRL, said, "BEIT has been exploring the limits of quantum algorithms, but we face the same obstacles when it comes to hardware performance. Fire Opal software opens up new frontiers in our research, bringing us closer to quantum advantages and achieving better results than classical algorithms in performing basic quantum algorithms. So, to some extent, this kind of software radically improves the hardware, achieving results that past industry benchmarks would have deemed impossible. ”
Grover quantum search algorithm test results, picture from Q-CTRL