Quantum Computing's Use Cases

When someone first learns about quantum computing, several questions usually come to mind. Of these, the first question is usually, “what is quantum computing?” Among the secondary others, at least one question usually concerns whether one needs to learn more about quantum computing or not. Someone may choose to learn quantum computing simply because someone might find it interesting. Still, business professionals rightly wonder whether or not there are applications that are relevant to their respective businesses. Business professionals require use cases to justify the time and expense of investing in any new technology, let alone quantum computing.

What is quantum computing and its applications?

Quantum computing can be thought of as a highly-specialized form of computation. Overly simplified, computers generally perform computation with their Central Processing Units or CPUs for short. However, specialized hardware exists for very specific classes of computation. The best-known example is the Graphics Processing Unit, or GPU, which is commonly used for editing videos, rendering 3D graphics, mining cryptocurrency, and training machine learning models. Another example is the Tensor Processing Unit, or TPU, which is used for very specific machine learning applications. Computer users who do not need such specialized computation do not need such specialized hardware.

Similarly, Quantum Processing Units, or QPUs, are believed to be useful for specific classes of computation. They will not replace CPUs any more than GPUs and TPUs have. And like GPUs and TPUs, not everyone will need a QPU. But also, like GPUs and TPUs, those who have a need for QPUs stand to benefit greatly by using them. A machine learning engineer can train models much faster with a GPU than with a CPU. A chemist will someday be able to simulate much larger molecules with a QPU than with a CPU or, for that matter, a supercomputer with many GPUs.

Solvay Conference participant Scott Aaronson sees the applications of quantum computing as five classes:

1.   Simulation of quantum physics and chemistry, including materials science

2.   Breaking cryptography

3.   Optimization

4.   Machine learning

5.   Proofs-of-concept

Some industries, such as the manufacturing industry, are exploring all of the above. Other industries, on the other hand, may only need a few. In any event, one major reason to be exploring quantum computing today is to not only potentially realize a future advantage over competitors but to at least not fall behind and someday experience a competitive disadvantage. What those competitive advantages and disadvantages might look like will vary from industry to industry.

What are quantum computing use cases?

Don’t let the five application classes of quantum computing give the false impression that quantum computers will have limited use cases. Those are just broad classifications. Within each of these classes, considerable research is being done. This article by no means lists all of them, partly because we can’t possibly know what every company and every university is researching. However, much of this research has been made public and continues to be made public, so the following is intended merely to broaden the scope of potential quantum computing use cases. Some of the current and potential future use cases for quantum computing follow.

Regarding the simulation of quantum physics and chemistry, including materials science, many industries are taking note. The automotive industry, for example, is interested in the potential to make better batteries, including EV batteries and solar cells, as well as carbon fixation, aka “better carbon capture and reduced carbon emission,” and even computational fluid dynamics (which is also of interest to the aerospace industry). The agriculture industry, for another example, is interested in such issues as desalination, “enhanced soil usage for higher crop yields”, fertilizer and other chemical reactions, greener fertilizer production, and sustainable development. Furthermore, the pharmaceutical industry is interested in chemical catalysts, molecular drug discovery beyond small molecules, protein folding, and simulating molecular structure.

The threat of breaking cryptography affects nearly everyone. Defensive cryptography and “quantum-safe” post-quantum cryptography are particularly interesting to the banking, financial, healthcare, and insurance industries. Where there's data being classically protected, there's the potential for future quantum attacks. In just the pharmaceutical industry, there's patient data, intellectual property, financial information, and more potentially vulnerable. Research areas include networking, random number generation, and Quantum Key Distribution (QKD). And, of course, governments worldwide are concerned about the impact on geopolitics, national security, and national defense.

Meanwhile, wherever there is a graph, a schedule, or a constraint, there is a need for optimization. Business development departments across all industries stand to benefit from ad optimization, loyalty rewards optimization, and product design. Manufacturers, wholesalers, warehouses, distributors, and retailers, as well as the transportation industry they all rely on, could benefit from better loading of cargo aircraft, logistic optimization, including shipping logistics, route optimization, scheduling optimization, and supply chain optimization. Furthermore, cities around the world may see improvements in airport security, infrastructure optimization, including power grid optimization, as well as traffic flow optimization, including bus route optimization.

Finally, machine learning interests all industries working with large amounts of data. Classification, for example, has medical and law enforcement applications, as well as utility assuring quality in manufacturing and predicting traffic patterns for autonomous vehicles. Natural Language Processing (NLP) is a powerful tool for retailers, utilizing artificial intelligence to analyze natural speech and recommend, up-sell, and cross-sell products. And weather forecasting, to mention just one more, applies to agriculture, air traffic, forestry, marine transit, and much more.

The proofs-of-concept classification has no specific use cases. The general idea is that quantum computers may be useful someday for demonstrating the feasibility of methods and ideas and so forth. In one sense, the Deutsch-Jozsa Algorithm can be considered an example. This textbook quantum computing algorithm was the first to demonstrate that quantum computers could perform certain classes of computation much faster than classical computers. The algorithm is in and of itself of no other value; it cannot be used to solve any real-world problems of any practical nature. However, it proved, in principle, that quantum computational advantage is possible.

What is the benefit of quantum computing?

One often-stated benefit of quantum computing is that quantum computers can perform certain computations much faster than the world’s most powerful supercomputers. In some cases, the proposed advantage is extreme: hundreds of seconds versus thousands of years.

With that latter example, classical computation becomes unfeasible. One famous example, often used by Dr. Robert Sutor of Classiq’s partner ColdQuanta, is that of a caffeine molecule. The caffeine molecule is not huge and is nowhere near the largest molecule scientists would like to simulate, but fully simulating this modest molecule would require a supercomputer constructed out of 1-10% of the mass of planet Earth. Since it is unfeasible to carve up such a huge chunk of planet Earth, or even the Moon, just to build a supercomputer, scientists are presently limited to performing partial simulations.

A quantum computer, on the other hand, would need only 160 fault-tolerant qubits to fully simulate a caffeine molecule. A full simulation is of much greater value than a partial simulation. And even though humanity is still quite far away from realizing a fault-tolerant 160-qubit quantum computer, it is far more practical to build such a device than it is to mine enough asteroids to build a supercomputer of sufficient size.

Another wonderful benefit of quantum computers garners insufficient public attention: it’s greener. By performing some calculations much faster than even a supercomputer, a quantum computer uses “a fraction of the energy used by classical computers.” In fact, quantum computing uses “a mere 0.002% of the energy of the supercomputer.” As you can see, quantum computing provides an opportunity to greatly “reduce energy consumption in data centers.”

What opportunities can you see with quantum computing?

Many of the prospective use cases of quantum computing don’t fit nicely into one of Professor Aaronson’s classifications. Just a few of these potential use cases include:

• Compose music
• Defensively develop and protect intellectual property through patents and publications
• Insurance against not adopting quantum computing (competitive disadvantage) 
• Quantum-inspired algorithms

Have you heard of any other use cases being considered for quantum computing? We guarantee there are more. In fact, this list is only the tip of the proverbial iceberg. What is your organization considering using quantum computers for?

How will quantum computers change the world?

From a certain point of view, quantum computers are changing the world now. First, engineering breakthroughs are being announced regularly. Companies like ColdQuanta are using lasers to ultracool atoms to nanoKelvins of degrees above absolute zero to be able to use them as qubits. And that’s just one example; imagine how all the engineering breakthroughs arising from the quantum computing industry will benefit the world.

Second, quantum physics is transitioning from theoretical to experimental. Using ColdQuanta as an example again, physicists worldwide can make and experiment with Bose-Einstein Condensates (BEC), also known as “quantum matter,” using their cloud-accessible Albert system. Although Albert is not a quantum computer, its younger cousin Hilbert will also leverage ultracold atom technology.

Furthermore, computer science is making great strides. Since Ewin Tang set the example with recommendation systems, researchers have been inspired by quantum algorithms to accelerate existing classical algorithms. This quantum-inspired approach reaps immediate benefits because classical algorithms can be implemented today. The challenge, as was the case after Ewin Tang’s breakthrough, is to design even more powerful quantum algorithms.

And finally, quantum computers are far more environmentally friendly than supercomputers. That estimate includes, by the way, the implementation of extreme refrigeration and all the power consumption that entails. However, some qubit technologies operate at room temperature and can shed the dilution fridge, possibly reducing energy consumption further.

Quantum computers are not going to replace personal computers. Countless applications will continue to run on current devices simply because it is more efficient to do so. That said, the use cases of quantum computing extend far beyond factoring numbers and unstructured search. The future of quantum computing is looking bright, in fact, for just about everyone.