Quantum Computing 101: Brilliant, Google, Microsoft Training Workers

Xconomy San Francisco — 

From Microsoft and IBM to Alphabet’s unit X and Canada’s D-Wave Systems, companies are racing to build powerful quantum computers that may solve problems beyond the capacity of the most sophisticated conventional processors, and do it much faster.

It’ll be some years before such uber-computers are robust and reliable enough for broad commercial use. But tech companies are already writing the novel kinds of software these revolutionary computers will need in order to operate—and businesses are even helping to train the workforce for a future era of “quantum speed-up.’’ For that training task, Microsoft and X, the cutting-edge research arm of Google parent company Alphabet, have partnered with a San Francisco-based educational technology company called Brilliant.

In a new course offered by Brilliant, students can learn the basics about quantum computing, write some quantum code, and give it a spin on a system that simulates the workings of a quantum computer—all using their smartphones if they like, says chief operating officer Eli Ross.

Advanced skill in quantum computing is rare—even more rare than expertise in other emerging fields where engineers are highly sought, such as in artificial intelligence disciplines like machine learning and deep learning. The global number of high-level researchers in quantum computing may be less than a thousand, the New York Times estimated in October. And that scarcity could stymie progress for companies in the field.

“To date, the tech talent shortage in quantum science has been a critical bottleneck to the industry’s progress,’’ the quantum software company Zapata Computing said as it announced a $21 million Series A funding round in April.

Brilliant and its two big partners hope to do something about that bottleneck.

Brilliant teaches STEM topics, from math fundamentals on up, by feeding online students with small, digestible bites of content along with exercises that encourage them to make immediate use of what they’ve learned.

Among Brilliant’s courses is an introduction to basic, classical computer science concepts, like the fact that each “bit’’ of information consists of only one of two numbers: zero or one. (A string of 8 bits, such as 0 1 0 1 1 0 1 0, makes up a byte, which codes for a single letter or other character, such as an “A” or a “&.’’)

In its new course, though, Brilliant is teaching students to set aside that traditional CompSci canon and grapple with the bizarro-world perspective of quantum computing, which looks at 0 and 1 and says, “Why choose?’’

In quantum computing, the units of information are known as quantum bits, or “qubits,’’ that can each stand for 0, or 1, or—brace yourself—they can be said to represent both 0 and 1 at the same time. This startling feat is achieved when quantum computer designers nudge the qubits into a state called superposition.

Tech innovators anticipate that computers based on such underpinnings could outperform conventional processors by exponential leaps when tackling certain complex problems. (Dive in here for a quick explainer from WIRED to find out how the quantum speed-up might result from this technology.) The technical challenges are huge, but companies are pouring resources into creating qubit chips, quantum software, and cloud-based services where enterprise-scale businesses can test out early quantum computers.

Not surprisingly, conventional software algorithms won’t do for these quantum computers. They’re based on quantum mechanics—which concerns the behavior of atoms, subatomic particles, and energy—-rather than on the classical physics that describes the familiar behavior of relatively large-scale things like tennis balls. (Subatomic particles such as electrons can exist in transient, in-between states, similar to superposition in a qubit.)

Brilliant’s class in quantum computing moves on to explain much more complex and counterintuitive concepts—like what happens when two qubits become entangled. Along with superposition, entanglement is a key technique in amplifying computing power. There’s a payoff for mastering all this weirdness, because it opens up a universe of strange possibilities. (Hint: Teleportation!)

The course introduces students to Microsoft’s Q# programming language, which was designed to work with the real quantum computers that have already been built. Though these machines are still at an experimental stage, and so far lack the capabilities needed to tackle the most challenging computational tasks, they are already serving as test beds for their inventors and for selected companies invited to explore their potential.

Developers of these powerhouse computers want more researchers and programmers to be up to speed on quantum computing—and standing ready to pose really big questions—as the technology matures and becomes more widely available. Quantum computers could deliver breakthroughs in fields from code-breaking to drug development and to the design of super-catalysts, advocates say. Their anticipated ability to bust through the encryption that shields sensitive data raises national security concerns—one of the reasons why quantum computing is an arena of intense international competition.

Cambridge, MA-based Zapata Computing is one of the US companies competing to make hires from the scanty global pool of quantum computing experts. Zapata’s work on a software platform for quantum computing is based on a collection of quantum algorithms it licensed from Harvard University. Zapata’s hiring struggles may have been worsened by restrictive US immigration policies, its CEO and co-founder Christopher Savoie told the New York Times late last year.

Zapata helped organize a campaign to bolster US education and research on quantum computing as a co-founder of the Quantum Industry Coalition—one of the groups that advocated for the passage of a new federal statute, the National Quantum Initiative Act, which became law in December. It sets up a multi-part initiative to support not only research in quantum computing, but also a raft of US educational programs in the field. Much of this 10-year initiative will involve universities and research centers, which can apply for millions of dollars in funding from federal grants.

Companies are already contributing to the training of experienced engineers through entry points such as Microsoft’s Quantum Development Kit.

But Brilliant and its partners Microsoft and X also provided students anywhere a quick way to get started in the field when the company first offered its Quantum Computing course to existing users in December.

Since then, 26,000 people have enrolled in the course, COO Ross says. Although Quantum Computing is classified as one of Brilliant’s advanced courses, students don’t need to be professional programmers to grasp the material. Brilliant suggests two of its courses as prerequisites: Linear Algebra and Computer Science Essentials.

Brilliant, founded in 2012, targets everyone from high school and college students to adults who just like to give their brains a workout. The company started offering courses such as “Probability’’ and “Physics of the Everyday’’ in late 2016. Now there are more than 40 courses, and subscribers can tap into them all for a yearly subscription of $120, or $10 a month. Other topics among the advanced courses include artificial neural networks and machine learning.

The Quantum Computing course is Brilliant’s first formal collaboration with industry partners, but the company has frequently worked with tech professionals on course design and writing, Ross says. Brilliant routinely makes the earliest unit of each of its courses available free online. But for now, it is providing open access to thirteen lessons of the 33-lesson Quantum Computing course for free.

After finishing the Quantum Computing course, students can proceed to take two of Brilliant’s related advanced courses: Quantum Objects and Computational Biology. The biology course covers the art of creating algorithms that can help predict the three-dimensional shape and structure of biological molecules from a knowledge of their DNA sequences. The structure of a biological molecule is key to understanding how well it may function as a drug or a catalyst.

Though the Quantum Computing course was quietly launched for Brilliant’s more than 8 million existing users in December, a recent Microsoft blogpost about the project was part of a more public debut.

Microsoft says its engineers worked with Brilliant to integrate its Q# quantum programming language into a coding environment that students can use through their browsers.

“For the first time, without installing a new development environment, programmers can try out quantum algorithms for themselves and learn how they can be applied to their own work,’’ the blogpost says.

The Q# language is also integrated with the classical programming language Python, giving Brilliant’s students a taste of the hybrid computing methods that will be used in the near-term period while quantum computing is still in its early stages of development. Classical computers could operate conventionally, but also call on quantum hardware to solve particularly hard problems, Microsoft says.

‘’By the end of this course, students will appreciate how a difficult classical problem can be translated into a quantum representation, and experiment with the reality of quantum computation,’’ the Microsoft blogpost states.

Brilliant’s other goal is to speed up the development of quantum computing by expanding the workforce of knowledgeable people.

“The course aims to grow the number of people who understand this field from (a) few thousands to the hundreds of thousands,’’ the company says.

Photo credit: Depositphotos