Wells Fargo prepares for the quantum leap

The financial institution explores the potential contributions from quantum computing and validates use cases in light of the forthcoming upheaval. As Wells Fargo CIO Chintan Mehta says, “We can not miss the possibilities of quantum.”

It was in 2019 that Wells Fargo discovered the potential of quantum computing. At the time, the finance company was exploring post-quantum cryptography and was already supporting research conducted by academic institutions in the field of artificial intelligence. When IBM began working with Quantum, Wells Fargo decided to explore its possibilities. “As we unpacked post-quantum cryptography approaches and what quantum had to offer in general, it became clear that quantum could be used for utility cases, including performing certain transactions exponentially more efficiently,” explained Chintan Mehta, CIO of Digital Technology and Innovation at Wells Fargo. The team also identified computational problems that classical computers are unable to solve within a reasonable time frame and that quantum techniques can potentially solve. “We had seen more opportunities to solve mathematical problems with the prospect of increasing productivity significantly,” recalled Mr. Mehta.

Today, Wells Fargo is collaborating with the MIT-IBM research group to explore and test mathematical calculations using quantum techniques. The experiments focus in particular on approaches to vector mathematics and generalized linear algebra. As an example of a use case, the CIO mentions the rapid recalculation of the price of a transaction ledger performed in parallel. It is more efficient by relying on the quantum ecosystem. Other application cases may be of interest to the financial industry, for example by exploiting data modeling capabilities to address the complex data structures on which fraud detection systems depend. When they fail, fraud detection mechanisms can delay customers’ onboarding by weeks. Quantum calculation should

significantly reduce treatment time. Mehta notes that Wells Fargo focuses on the utility aspects of quantum computing. “We participate in quantum research to help validate use cases in the financial sector, which will benefit. We do not do pure basic research, for example, we do not try to build a physical quantum infrastructure,” DSI clarified.

The promises of quantum

Traditional data processing relies on binary arithmetic, easily processed by chips. Conventional computers have made progress by plugging more transistors into increasingly smaller integrated circuits. Quantum computing disrupts the rules of the game by using what is called a qubit (the quantum equivalent of a bit), which is capable of adopting more states than binary 0 or 1. The power of quantum systems grows exponentially, meaning that a theoretical system of 200 qubits will be 2,200 times more powerful than a system of 100 qubits.

Quantum computers could therefore tackle problems that their traditional counterparts simply do not have the capacity to solve, such as complex multivariate problems based on probabilities and modeling of hypothetical scenarios. They could help solve problems like those with driver assistance apps when it comes to choosing alternatives to congested routes that then create new traffic jams. Traffic optimization by quantum calculation makes it possible to solve it by calculating all the possibilities at the same time. A large neural network, the basis of many advanced computations across industries, relies on linear algebra calculations to train billions of nodes. “Quantum computing is accelerating this process exponentially,” Mehta said. “It will only take a few minutes to set up a global network compared to several days today,” he added.

Known unknown

“Despite the promise of quantum technology, one of the challenges of working with this technology is the many unknowns,” Mehta noted. “There’s still a huge gap between the quantum properties we imagine and what it actually can, especially when it comes to solving higher-order mathematical operations,” he said, but explained. Adding that at the moment, “nothing is known about the stability of these operations, or about the expected repeatability of a computer construction, all this is fundamentally absent in the current quantum. You can run a calculation several times and potentially get different answers every time.” This is a source of concern. “We are in a phase where simulation is much more advanced than the actual physical quantum computer,” Mehta said.

The other uncertain is whether the technology will evolve as expected. Mr. Mehta compares quantum technology to artificial intelligence: “In the case of AI, we knew the technology worked, but we did not know if a user could be sure that a specific model would work in conjunction with his business,” he said. “In the case of quantum computing, the likelihood of failure is higher because nothing has been proven and there is no common baseline for measuring success,” he said. What would success look like for Wells Fargo? “It will be more about reaching major milestones than getting very specific results,” Mehta said. “Will we be able to do more simulations? Fourier transforms? Differential calculus? There will be more steps before going from very basic discrete mathematics to more sophisticated calculations,” he added. The ultimate goal of Wells Fargo is to have “a good library of mathematical possibilities from which we can build a use case,” said the CIO.

Prepare the terrain

Because the goal of quantum success is consistently marking a list, Mehta advises companies to commit to a small but long-term program. “Companies can not spend all their money on it, but they can research consistently,” he advised. The fallout is huge, but will only be felt for a long time. James Sanders, an analyst at 451 Research, also advises being patient and playing long-term. Sanders, who is part of 451 Research’s S&P Global Market Intelligence team, says companies need to start building software now. “Now is the time to start investigating potential issues that the company cannot solve today due to lack of computing power,” Sanders said. “It’s time to take stock of the problems that could be solved by quantum computers,” he added.

This inventory work is the first of five tips for companies formulated by SSII Capgemini so they are preparing to take advantage of the quantum advantage. “This assessment is crucial. It requires computer scientists and business experts to identify industrial and business problems that cannot be solved with traditional computers and that quantum computers could solve,” said Satya Sachdeva, vice president of insights and data at Capgemini. a small team of experts, translate the most interesting use cases into small quantum experiments, enter into long-term partnerships with technology providers to overcome technical obstacles and develop a long-term strategy to increase their skills, according to Lian Jye Su, lead analyst at global technology firm ABI Research “Public cloud players are spoiled for choice.” Public cloud players like Alibaba, Amazon, Google and IBM provide services to enable researchers to run quantum applications and experiments remotely, “said Su. Developers can now build quantum applications using IBM’s Qiskit, Google’s Cirq, Amazon Braket and others, s whether all are “open source libraries designed to optimize quantum circuits for quantum classical or only quantum, including machine learning,” Su said. “All of these services are available online.”

The Pioneer’s Advantage

Experts believe that quantum data processing and classical data processing will exist side by side in the long run, and that some processes will be handed over to quantum data processing before being reintegrated into classical data processing. Capgemini’s Mr Sachdeva says now is a good time for companies to get started. “Early adopters will have an advantage over others, and companies that procrastinate for too long will lose that advantage,” he said. As for Wells Fargo, the financial institution is slowly maintaining its quantum program. According to Mehta, CIOs must ensure that all advanced research projects are carried out in parallel. “We must continue to improve other non-quantum capabilities, whether it is specialized cloud-based infrastructure or other technologies, in parallel,” Mehta recommended. “When quantum computers materialize and computer properties converge, this step will be both very disruptive but very positive.” It’s time to prepare for this disruption, as Wells Fargo does. “We think quantum computers are where classical computing was 30 or 40 years ago,” Mehta said. “That said, it will evolve much faster than anything else we’ve experienced before. We can not refrain from such an upheaval. This technology is crucial for the future.”

Leave a Comment