Skip to main content

Why Do We Want a Quantum Computer?

news_021620_plenary_full
Complex problems from traffic flow to pollution could benefit from a quantum computer, according to Microsoft engineer Krysta Svore. | Robb Cohen Photography & Video

Microsoft engineer Krysta Svore wants a quantum computer to solve the FeMoco problem, but that’s just the start of what she thinks a quantum computer can do.

The FeMoco molecule is found in tiny organisms that help fix nitrogen in the soil to provide natural fertilizer to plants. If scientists could better understand FeMoco chemistry and “if we could mimic what these organisms are doing in the soil industrially, we could help reduce 3% to 5% of the world’s natural gas consumption that goes to artificial fertilizer production,” Svore explained in her plenary address at the American Association for the Advancement of Science’s Annual Meeting.

Understanding even the lowest energy state of FeMoco, however, is beyond the capabilities of even the world’s fastest supercomputer. It would take that computer longer than the lifetime of the universe to make the necessary calculations, where a quantum computer may someday be able to perform the calculations in days or weeks, she said.

Digital computers, from the laptop to the supercomputer, operate using strings of bits—discrete units of data represented by 0s and 1s. During a particular operation, the computations have to take place one string of bits at a time.

The unit of information in quantum computer is a qubit, Svore explained. Qubits aren’t just a 0 or a 1, but can represent numerous combinations of 0 and 1 at the same time, by taking advantage of the strange properties of quantum mechanics. This allows a quantum computer to make numerous computations simultaneously, like “computing on all the bit strings at the same time,” Svore said.

Some of the world’s most technical challenges, including encrypting sensitive data, building natural language systems in computers, and figuring out the most cost-effective and fuel-efficient routes for delivery vehicles, are among the problems that could be solved with quantum computers, she said.

Svore explained how researchers around the world are building a full ecosystem of quantum computing, from the hardware to the programming languages to the open-source educational resources necessary to make quantum computing viable.

The first step is to build a qubit, which can be made from a variety of materials including metallic quantum dots and nanowires. Because a qubit only retains its special quantum state if it is not disturbed before it “reads out” its calculation, researchers also must engineer a special protective system around it. In the case of Microsoft’s quantum project, the protection comes in the form of intense cooling—down to millikelvin temperatures or near absolute zero—as cold as the outer reaches of space.

Since engineers don’t want to program in these frigid temperatures, Svore joked, the quantum ecosystem also needs ways to control qubit operations from room-temperature digital computers. The end result is a kind of hybrid system that includes new languages for writing quantum algorithms and simulators to test how well a quantum program will run.

Some of those tests are already being run on the FeMoco problem. A quantum algorithm written in 2012 for FeMoco would have required 30,000 years of calculation. With steps to test and improve the algorithm, Svore said, researchers managed to bring that run time down to one and a half days by 2015.

Microsoft’s quantum team is already working on other real-world applications for quantum computing. Researchers at Case Western Reserve University use quantum algorithms to transform MRI scans for cancer, allowing the scans to be performed three times faster and to improve their quality by 30%. In practice, this could mean that some children won’t need to be sedated to stay still for the length of an MRI, and physicians could track the success of chemotherapy at the earliest stages of treatment, said Svore.

Svore said applications like the MRI project at Case Western are “just scratching the surface. And we need many, many more software engineers and algorithm designers helping develop new quantum algorithms.”

Before the plenary address, AAAS CEO Sudip Parikh presented two AAAS awards: the Science Diplomacy Award and the Scientific Freedom & Responsibility Award. AAAS President-Elect Claire Fraser also announced the theme of the 2021 AAAS Annual Meeting, “Understanding Dynamic Ecosystems,” to be held Feb. 11-14 in Phoenix, Arizona.