Quantum computing a new industrial revolution
Investors active at the turn of the century will recall two particular technology lessons. One was a skepticism about the industry's unprofitable business. The o阿ther was the overwhelming fear of the proposed Y2K problem, when it was feared that an anomaly in the calendar program would crash the system and cause planes to fall out of the sky. As it turned out, thanks in part to a lot of work behind the scenes, this sort of thing didn't happen.
So, should you discard the hype of an arms race in quantum computing technology? The U.S. and China are pouring billions of dollars into the field with an eye toward using its problem-solving potential to break each other's security encryption, and European countries are following suit. But will this really translate into something tangible and valuable for investors?
Consider that companies such as BMW, Goldman Sachs, JP Morgan Chase, Roche, LG Electronics, ExxonMobil and Mitsubishi Chemical are seriously working on their own quantum computing solutions, and the skepticism may disappear. Potential applications in a wide range of industries increase the likelihood of adoption.
Keen observers will also note Alphabet's divestiture of its Sandbox AQ division (A for artificial intelligence and Q for quantum computing) in March 2022. Achieving a nine-figure funding round suggests that the project is entering a new phase of growth, with venture capitalists ready to fund it, even as public markets with an eye on short-term gain shy away from it.
For private investors, the options are also growing. While there are few pure quantum computing companies in the public market, there have been a number of U.S. listings through special purpose acquisition companies (SPACs). Rigetti and Ion-Q have gone public in this way since the fourth quarter of last year, and D-Wave completed a SPAC merger last month to go public as D-Wave Quantum.
Established companies with strength in other established revenue channels could also benefit. Commercially viable solutions could be transformative for IBM, which is at the forefront of quantum computing.
There are also adopter companies that foresee a competitive advantage in mastering this technology, which promises to provide a new and faster way to overcome computing challenges. Making a market assessment of the opportunity is difficult, and Gartner Research consultant Chirag Dekate likens business progress to the first five minutes of a marathon. While annual spending by potential end-user companies has risen steadily over the past five years, it goes without saying that the $3 million average Dekate talks about must grow further to show that quantum is indeed a game-changer.
Quantum computing is not only a faster way of computing, it is also a different way of computing. While classical computers perform linear calculations using binary numbers (bits) as the standard unit of information, quantum bits solve problems by exploiting the properties of space and time. Scientists are particularly interested in three of them: superposition, entanglement and interference.
These properties allow quantum computers to solve certain types of problems. Simply put, this is much more efficient than flipping a coin multiple times and then inserting the information into more calculations. Thus, the application of quantum technology could be a revelation for any industry where solving probability problems is important. This is true for many industries.
Not surprisingly, a specific segment of the financial services industry is very promising to be a part of the quantum vanguard.
Correctly estimating risk parameters has always been a challenge for portfolio managers and investors, let alone banks whose models were imperfect after the onset of the global financial crisis. prior to 2008, many financial models relied on value-at-risk (VaR), a measure that had an inherent flaw in assuming that daily asset returns were normally distributed.
While VaR calculated the average loss on the worst 5%, 1% or even 0.5% of days, it grossly underestimated the likelihood and magnitude of the worst days. Progress has been made in risk modeling since the financial crisis, but quantum computing may lead to even greater advances.
For example, Monte Carlo simulations (mathematical models that rely on the flexibility of random number inputs) could include exponentially more variables if quantum bits were used to accurately calculate the probability of outcomes. This means that a more complete picture of potential causes and effects can be captured when estimating risk. It can even help model the impact of climate change on a portfolio.
Other industries that are directly in the spotlight for the clean energy transition are also looking for uses for quantum computing. Oil and gas giant ExxonMobil is working with IBM to specifically use quantum algorithms to solve the challenge of transporting liquefied natural gas (LNG), a problem associated with the energy crisis and the war between Russia and Ukraine.
Handling delivery logistics requires considering the location of every ship for every day of the year and millions of careful decisions. Add variables such as weather and demand fluctuations, and the variables involved in this problem increase to billions or even trillions. Quantum computing's ability to process and understand these small, complex and interconnected variables has the potential to make supply chains smoother and more efficient.
Automaker BMW is using the testing capabilities offered by quantum computers to help solve its complex supply chain and engineering challenges. Pharmaceutical companies such as Roche are using the technology to evaluate the impact of millions of tiny variables on experiments. Battery and appliance maker LG is also testing and developing their products.
The intersection between quantum computing and AI learning is particularly interesting: "What's most exciting is the potential for researchers to gain new inspiration from the fundamentally different ways quantum computing operates and apply those insights to classical AI that can be leveraged in the short term," Sandbox AQ CEO Jack Hidary said.
For successful quantum technology to revolutionize cybersecurity, Hidary believes machine learning will be critical for systems to "adapt to emerging threats and implement the best, most appropriate cybersecurity algorithms in real time.
The ability of quantum computers to test thousands of permutations simultaneously, combined with the ability of machine learning to identify patterns, could also be applied to drug testing. Ultimately, improved prediction of negative or dangerous outcomes could safely accelerate drug development and reduce costs.
02Superconducting and ion trap quantum bits for different problems
Because of the wide variety of uses and specific end-user needs, the most appropriate quantum computing technologies vary depending on the problem being solved, including superconductivity, semiconductors, photonics, and ion trap technology.
IBM's Dr. Olivia Lanes said, "IBM is focused on building processors out of superconductors and what we call Transmon quantum bits." Unlike semiconductors, there are no supply issues: many of the leading superconductor manufacturers are located in the United States or Japan.
"Transmon has proven to be very good at mutual coupling," Lanes said, adding that the technology allows for fast control. For superconductor technology, the computational step between two quantum bits is very fast (about a few nanoseconds). But Tony Uttley, chief operating officer and president of Quantinuum, a quantum computing company controlled by Honeywell, says this method is not as good as the ion trap method at keeping information stable. But on the other hand, the latter is slower. "There are trade-offs among all of these technologies."
Deciding which method to use may depend on the number of quantum bits that need to communicate with each other. For example, Uttley said, certain types of chemical experiments require multiple quantum bits for arbitrary communication, which may be better suited to an ion trap solution, where the quantum bit connections can be maintained for about 12 minutes. For problems where the quantum bits need to interact primarily with their nearest "neighbors," superconductivity may be better.
Achieving the optimal speed and maintaining the stability of the quantum state for the longest time (this is called coherence, decoherence is the return to the ground state) is an ongoing challenge. They are as important as the number of quantum bits. "You could have a million really bad quantum bits and it still wouldn't be of any use. What matters is speed, quality and scale, and in order to achieve quantum dominance, we have to make all three very powerful," Lanes said.
Also to be emphasized is error correction, as it is critical to maintaining the quantum state and generating the highest quality and most useful quantum bits. lanes added: "All quantum bits have a T1 lifetime, which is the time before they decoherence back to the ground state."
In most cases, certainly where superconductivity is the underlying technology, the T1 time is on the order of microseconds. "If you need to execute your algorithm, your computation time is going to be faster than the T1 time. you can't execute complex algorithms in that amount of time, so you have to introduce error correction mechanisms," Lanes said. Once error correction is in place and working, it doesn't necessarily matter how long that information can be kept, because when the error happens, we can find it, explain it and fix it."
Another problem is so-called leakage: when quantum bits are excited by too much energy and "leak" to higher energy states, making it harder to correct errors. "When we scale the processor, we need to make sure we have a low leakage rate and associated error rate," Lanes added.
Reference link：
https://www.investorschronicle.co.uk/news/2022/09/05/quantum-computing-a-new-industrial-revolution/
