In an article published in Quantum information about nature, EPFL professor Giuseppe Carleo and Matija Medvidovi ?, a graduate student at Columbia University and the Flatiron Institute in New York, have found a way to run a complex quantum computing algorithm on traditional computers instead of ‘quantum computers.
The specific “quantum software” they envision is known as Quantum approximate optimization algorithm (QAOA) and is used to solve classical optimization problems in mathematics; it is essentially a way of choosing the best solution to a problem from a set of possible solutions. “There is a lot of interest in understanding what problems can be solved efficiently by a quantum computer, and QAOA is one of the more prominent candidates,” Carleo said.
Ultimately, QAOA is meant to help us on our way to the famous “quantum acceleration,” the predicted increase in processing speed that we can achieve with quantum computers instead of conventional computers. Naturally, QAOA has a number of supporters, including Google, who have their sights set on quantum technologies and computing in the near future: in 2019, they created Sycamore, a 53-qubit quantum processor, and used it. to perform a task that he estimated. It would take about 10,000 years for a conventional state-of-the-art supercomputer. Sycamore completed the same task in 200 seconds.
“But the barrier of ‘quantum acceleration’ is almost rigid and it is continually being reshaped by new research, also thanks to advances in the development of more efficient classical algorithms,” explains Carleo.
In their study, Carleo and Medvidovi? answer a key open question in the field: can algorithms running on current and short-term quantum computers offer a significant advantage over classical algorithms for tasks of practical interest? “If we are to answer this question, we must first understand the limits of classical computing in the simulation of quantum systems,” Carleo explains. This is all the more important as the current generation of quantum processors operate in a regime where they make errors when executing quantum “software” and can therefore only execute algorithms of limited complexity.
Using conventional computers, the two researchers developed a method capable of roughly simulating the behavior of a special class of algorithms called variational quantum algorithms, which are means of determining the lowest energy state. , or “ground state” of a quantum system. QAOA is an important example of such a family of quantum algorithms, which researchers believe are among the most promising candidates for a “quantum advantage” in quantum computers in the short term.
The approach is based on the idea that modern machine learning tools, for example those used in learning complex games like Go, can also be used to learn and mimic the inner workings of a quantum computer. The key tool for these simulations is Neural Network Quantum States, an artificial neural network that Carleo developed in 2016 with Matthias Troyer, and which is now used for the first time to simulate QAOA. The results are considered to be in the field of quantum computing and constitute a new benchmark for the future development of quantum hardware.
“Our work shows that the QAOA that you can run on current and short-term quantum computers can also be simulated, with good precision, on a conventional computer,” Carleo explains. “However, this does not mean that all useful quantum algorithms that can be executed on short-term quantum processors can be emulated in the classical way. In fact, we hope that our approach will serve as a guide to design new quantum algorithms at the same time. both useful and difficult. simulate for conventional computers. “
Source of the story:
Material provided by Federal Institute of Technology in Lausanne. Original written by Nik Papageorgiou. Note: Content can be changed for style and length.