There's still a long way to go in making these very complex computers work, but even without diving into the details, we can describe the types of things that quantum computing will be very, very good at. And we can give a general idea as to which of these improvements might be available sooner rather than later.
A quantum computing processor from IBM
Thinking in triplicateThere are three broad categories of quantum computing applications. It’s useful to examine each task you’re trying to accomplish from all three of these viewpoints. Applying quantum computing to real-world problems is a creative task, especially in these early days, and using multiple viewpoints can only be helpful.
Here are the three approaches:
- Simulation: In simulation, qubits — trapped bits of coherent matter — mimic other coherent matter, such as the individual atoms within a molecule that might become a medically useful drug. Simulation is arguably the most natural fit for quantum computing because quantum mechanics is what governs the laws of, well, nature.
- Optimization: A group of qubits can be used as a kind of computational furnace that can be guided into yielding a very good — but not necessarily perfect — solution to a problem. The result might be the right answer, or it may instead be something close to that. (A very good solution to a route-planning or investing problem might save, or make, you a lot of money, even if it isn’t the best possible answer.)
- Calculation: This approach is, conceptually, the most like the classical computing problem-solving we’re all used to. In calculation, qubits are combined into logic gates, making up a universal computer. When used as logic gates, qubits can solve any imaginable problem, and a quantum universal computer can solve some important problems far faster than today’s computers — which also fit the “universal computer” description — but grind to a near-halt for some problems.
Both the features used in machine learning and the operations against the Bloch sphere used for manipulating the qubits of gate-based quantum computers are stated as vectors, so the calculation approach is readily used for machine learning. (Although optimization can be used for machine learning as well.)
Algorithms can be grouped into these same three categories, which helps spotting areas where algorithms can be extended to accomplish additional goals. Importantly, the same quantum algorithm can underpin several different applications; for example, the algorithm that powers a financial portfolio optimization application might also underpin a separate application for route optimization.
Also, the categories of applications can overlap; for instance, if you use optimization to come up with better and better answers, you may at some point come up with the exact answer, just as if you used calculation. (For instance, using optimization to find the prime factors of a large prime number, just like Shor’s algorithm, which belongs in the calculation category.) But the categories are useful for understanding the current state of quantum computing and anticipating what progress we might expect in the near future.
Big potential for quantum computingThere are several areas in which quantum computing could far exceed the abilities of classical computing. Following, are summaries of some of these.
CryptographyQuantum cryptography is “the straw that stirs the drink” in quantum computing — a phrase first attributed to baseball great Reggie Jackson, who was working in an entirely different field (right field, to be precise).
The current, fervent interest in quantum computing began in 1994 with the publication of Shor’s algorithm, which is one of the few quantum algorithms that has been proven, at this early point, to have the potential for exponential speedup. However, Shor’s algorithm will be able to do useful work only when it’s run on quantum computers far more powerful than those available today.
Quantum computing has the potential to break the most common encryption methods used to secure digital communication today, such as RSA and ECC, which protect emails, bank information, the web, and more. These encryption methods rely on the difficulty of factoring large integers and the difficulty of computing discrete logarithms, respectively.
Quantum computers can perform these operations exponentially faster than classical computers, making them a threat to traditional encryption methods. Quantum algorithms have been proposed for key exchange, digital signatures, and encryption, which are the building blocks of secure communication.
Search algorithmsSearch algorithms have been an important area of research in computer science for decades. Real-world examples of the use of quantum algorithms for search include optimization problems in internet search, finance, logistics, and transportation.
For example, the use of quantum algorithms for portfolio optimization will help financial analysts find the optimal investment strategy for a given portfolio in a fraction of the time required by classical algorithms. (Using quantum algorithms to optimize your portfolio works especially well if you have a quantum computer and the other investors don’t.)
With the exponential growth of data, several algorithmic challenges need to be addressed. One of the biggest challenges is finding an optimal solution in a reasonable amount of time, which is where quantum algorithms come into play.
One of the earliest, best-known, and most promising quantum algorithms is Grover's algorithm, used for searching an unsorted database and for a wide range of other purposes as well.
For more details on these and other possible applications for quantum computing, check out our book Quantum Computing For Dummies.
Financial industry applicationsQuantum computing is starting to make waves in the financial industry, with many companies turning to this new technology in an effort to improve their operations and gain a competitive edge. Today, quantum algorithms and applications are being explored by a variety of financial companies for uses including portfolio optimization, risk management, and fraud detection.
Goldman Sachs, a leading investment bank, and several other banks are working to develop quantum algorithms for portfolio optimization; “the vampire squid,” as Goldman Sachs is sometimes called, has shown promising results in improving investment returns. By utilizing the processing power of quantum computing, this portfolio optimization effectively analyzes vast amounts of data and identifies investment opportunities that traditional algorithms might overlook, leading to more informed investment decisions.
With the capability to simultaneously perform multiple calculations, quantum algorithms can help financial institutions make more informed decisions while minimizing risk and maximizing returns.
Insurance risk analysis & fraud detectionOne area where quantum algorithms may be particularly useful in the insurance industry is in risk analysis. Insurance companies use risk analysis to determine the likelihood of a particular event occurring and the potential costs associated with that event.
Quantum algorithms could greatly enhance this process by allowing for more complex calculations to be performed in a shorter amount of time. This, in turn, would allow insurance companies to better assess risk and set more accurate premiums.
Another area where quantum algorithms could be beneficial in the insurance industry is in fraud detection. Fraudulent claims cost insurance companies billions of dollars each year. Detecting and preventing fraud is a top priority for many insurers. Quantum algorithms could help insurers more effectively identify fraudulent claims by analyzing large amounts of data and detecting patterns that might be difficult to spot using traditional methods.
LogisticsThe logistics industry is constantly seeking ways to optimize its supply chain processes, and one of the latest innovations that has emerged is the use of quantum algorithms.
Given the intricacies involved in supply chain optimization, quantum algorithms have the potential to be highly effective in this domain. They can facilitate the analysis of large data sets, optimize shipping routes, reduce transportation costs, and increase overall operational efficiency.
One easy-to-understand example of the power of logistics is the daily route planning used by delivery company UPS. They rather famously train their drivers, and design their routes, to almost always avoid turning left.
This is not some kind of political statement, but rather the result of the long waits that drivers of all vehicles sometimes suffer in getting the opportunity to safely make a left turn. By avoiding them, UPS drivers save time and money. (And might even avoid a few bent fenders along the way.)
Medical scienceOne of the most promising applications of quantum algorithms in medical science is in modeling the workings of the human body at the molecular level. Quantum computers can succeed here where classical computers fall short.
One real-world example of the use of quantum algorithms is the work being done by researchers at the University of Toronto. They have used quantum algorithms to simulate the behavior of a protein involved in the development of cancer. By doing so, they were able to identify a potential drug candidate that could inhibit the protein's activity, potentially leading to new cancer treatments.
Another area where quantum algorithms are showing promise is in medical imaging. MRI scans, for example, produce vast amounts of data that must be processed and analyzed to produce images of the body. Classical computers can struggle with this task, but quantum algorithms can handle it much more efficiently, which could lead to faster and more accurate diagnoses, as well as more effective treatments.
Finally, quantum algorithms are used also to improve our understanding of biological systems. By simulating the behavior of complex biological systems, researchers can gain new insights into how they work and develop new treatments for diseases.
PharmaceuticalsThe process of developing new drugs is incredibly time-consuming and expensive, with many potential candidates failing in clinical trials. However, quantum algorithms can simulate the behavior of molecules at a level of detail that's impossible for classical computers.
The effectiveness of quantum computers for this purpose means that researchers will be able to more accurately predict the effectiveness of different compounds, potentially leading to faster and more successful drug development.
One of the quantum algorithms being tried for drug discovery is the variational quantum eigensolver (VQE). This algorithm is used to determine the ground state energy of molecules, which is a critical factor in drug design.
The VQE algorithm uses a hybrid approach that combines classical and quantum computing to solve complex problems. It's particularly useful in drug discovery because it can accurately predict the molecular structure of compounds and their interactions with target proteins.
Another quantum algorithm that has gained traction in drug discovery is the QAOA algorithm we mentioned previously. It solves optimization problems, which are common in drug discovery. The QAOA algorithm uses a series of quantum gates to optimize the energy landscape of molecules, which helps researchers identify the most promising drug candidates.
Addressing climate change
Climate change is a looming crisis that requires innovative solutions. The use of quantum computing and quantum algorithms could be one such solution. These technologies can help us better understand climate patterns and predict future climate changes with greater accuracy.
By simulating complex systems and performing calculations at a much faster rate, quantum algorithms could help us identify ways to reduce carbon emissions, trap carbon from manufacturing processes or in ambient air, and develop more efficient renewable energy sources.