The field of artificial intelligence (AI) has brought to light several dilemmas that society faces regarding the friction between ethics and liberal values ​​and how they relate to the advancement of computer technology.

Although the The prospect of quantum superiority, where it would be impossible to simulate algorithms running on a quantum computer with a classical computer, raises serious questions about how one can prove that the quantum computer produces the correct results.

Given the problem space, given the multitude of parameters that a quantum computer can use to find an answer, how can a mere human determine whether the processing makes sense? “If a quantum computer can efficiently solve a problem, can it efficiently convince an observer that it is right?” Says Marc Carrel-Billiard, Global Technology Innovation Lead at Accenture.

Researchers are developing a better understanding of where Quantum computing can be used. Heike Riel, IBM Fellow, Head of Science and Technology and Head of IBM Research Quantum Europe at IBM Research, says, “It’s not about the beauty of technology. We want to create value – it’s a journey, develop the technology, find the sweet spot for early application, see and demonstrate value, then we add hardware and software. ”For example, Eon Energy recently joined the IBM Quantum Network . Energy sources were few in the past, but as the world moves to more green sources such as sun and wind, there are now many more sources to generate energy from. Quantum computing could help distribution grids to fulfill a much broader range of tasks, especially if in the future many smaller companies and households have their own photovoltaic (PV) systems or electric cars through initiatives such as Eons. Feeding energy into the grid Vehicle-to-Grid project (V2G).

In this project, batteries from electric vehicles are connected to the distribution network as a flexible storage medium. In this way, fluctuations in the generation of renewable energies can be balanced out. With the help of quantum computing, these processes could be controlled more efficiently and effectively.

Riel says: “All of these sources have different dependencies and therefore predictions become more complex. We need to optimize the system and we want to do it in real time. The complexity increases exponentially with the number of parameters and that becomes difficult to solve with classical computing. ”

Bob Coecke, theoretical physicist and senior scientist at Cambridge Quantum Computing, points out that atoms and molecules are subject to the laws of quantum mechanics , so their behavior should be modeled on a quantum computer. “The simulation of physical things is exponentially expensive due to the structure of quantum mechanics,” he says. “These things are quantum – they want to live on a quantum computer and it is artificial for them to live on a classic computer.” to integrate classical computers.

Modeling the behavior of atoms and molecules was the driving force behind the use of quantum computers to simulate new materials. Last August, Nicholas Rubin and Charles Neill, researchers at Google AI Quantum, wrote a blog discussing an experiment to create a complex chemical simulation using a Hartree-Fock model from computational physics.

“One accurate computational prediction of chemical processes based on the quantum mechanical laws that govern them is a tool that can open new frontiers in chemistry and improve a variety of industries, ”the researchers write. But on the blog, they admitted that algorithms to simulate chemistry on short-term quantum devices must take into account errors that occur on quantum computers.

Similar to how classic neural networks can tolerate imperfections in data, in their experiment the couple is trying the quantum processor – Equivalent of a neural network called a Variational Quantum Self Solver (VQE) to optimize the parameters of a quantum circuit to make loud quantum logic.

While the ability to simulate a complex chemical process on a quantum computer is amazing, scientists understand how the interactions of subatomic particles lead to a certain result. It can be written down and calculated using a chemical equation.

Riel at IBM says that as long as the number of qubits is small enough, results can be simulated on a classic computer. She says IBM is working to understand how noise affects the system by giving false results. As IBM continues the next stage of its quantum computing roadmap with a 128-qubit quantum system, “we want to demonstrate error correction and are working on verifying the results,” adds Riel.

Mark Mattingley-Scott, Managing Director of Europe at Quantum Brilliance explains the challenges: “One of the paradoxes of quantum computing is that once we get to the point of quantum utility, i. i.e., a quantum algorithm performing calculations with a speed and precision that are not possible on a classic computer, it becomes impossible to check the correctness of the results directly. We can check the correctness of the method with smaller versions of the same problem – which we do every day with classical algorithms – but there will be no way to actually check this. ”

Since quantum computing is inherently non-deterministic, Mattingley-Scott suggests that the results it produces are based on probabilities. “A quantum algorithm works by using quantum mechanics to constructively reinforce the ‘right’ answer and destructively suppress the ‘wrong’ answers,” he says. “So there will always be a certain amount of uncertainty. The validation of a quantum computer with a classical computer will only be possible on a methodological level, not on the level of the actual data. ”

Coecke from Cambridge Quantum Computing believes, however, that the principle of compositionality and category theory can help to understand what actually goes on in a quantum computer. Compositionality uses the idea of ​​thinking a problem both top-down and bottom-up. “All math works from the bottom up,” he says. “We are using a new form of compositionality in which the ‘whole’ defines the parts.”

Traditionally, in computer science, a complex system is defined by building and testing smaller, known parts that are integrated when they are integrated are behaving as defined by the sum of the parts. “Category theory is about how something relates to a larger picture,” says Coeck. “It’s an important part of computer science.”

He adds that category theory is used to structure how programs fit together. In this way, flowcharts are used to illustrate what a program does or a circuit diagram shows how a battery and switch are connected to a lightbulb.

The concept that a picture is worth a thousand words is explored by Coecke in a book he wrote together with Aleks Kissinger, “Picturingquantumprocesses”. While the target audience for this book is students, Coecke also publishes a book aimed at teenagers that shows that almost everyone can understand quantum computing.

By and large, the book is about the idea of ​​using a composition tool to break big problems down into small parts. According to Coecke, these small building blocks can be put together in an understandable way and, above all, all components can be checked.

Back to chemistry, says Michael Biercuk, CEO and founder of Q-CTRL: designing a particular set of target properties, in part because of the limitations of our computational modeling skills. However, it is easy to measure the properties of a candidate molecule using this list. In the case of molecular structure or chemical dynamics calculated on a quantum computer, we may not be able to do a classic validation simulation, but we can generally do a real chemistry experiment to validate the results. ”Similarly, Mattingley-Scott believes of Quantum Brilliance that one possible role for quantum computers is to provide dramatically accelerated performance. “It might be able to perform sensitivity analysis of the solutions to problems to determine what if and provide answers based on that,” he says. “The ability to do this is kind of the holy grail of quantum computing, and such an ability would revolutionize industry and society.”

As the world moves towards quantum supremacy, the experts at Computer Weekly agree that it is becoming increasingly difficult to prove that the output of a quantum computer is correct. Coecke’s pictorial top-down approach with verifiable building blocks could be a way of developing complex quantum algorithms. The idea of ​​verification through real-world experiments, as Mattingley-Scott suggests, could become the quantum equivalent of the integration test in classical computing.

But as Riel from IBM points out, quantum computing is one of many approaches that software developers must take to complex problems to tackle. “If you have a problem to solve in optimization, you don’t care how it is done as long as it is solved the fastest and most efficiently,” she says. “You don’t want to worry about which computer is being used.”

From IBM’s perspective, a complex computational problem can require different building blocks, with some parts being processed using traditional computing while others using quantum computing. Riel adds, “You need developers who understand quantum computing to develop the quantum primitives that implement algorithms. You then need a model builder who doesn’t need to understand the depth of quantum computing but is able to describe the problem and use the best solver application. The model developer shouldn’t bother himself with quantum knowledge. ”

Although quantum dominance may still be a long way off, companies today are starting to use real-world quantum processing to solve difficult problems. For example, pharmaceutical and materials science companies use a variety of computationally intensive methods to check molecular matches and predict the positive effects of potential therapeutic approaches while reducing negative side effects.

Researchers at Accenture Labs recently worked with Biogen to identify the quantum-activated optimization techniques that are most beneficial to the company. Accenture’s Carrel Billiard says such optimizations can be tested. Accenture is also working with clients in the financial sector to evaluate how existing algorithms run on a quantum computer. These are algorithms that are already running on classic computers and therefore their results have already been verified.

Similar to how Coecke from Cambridge Quantum breaks down a problem into verifiable parts, Carrel-Billiards team at Accenture is working on certain problems To map groups of mathematical primitives. These primitives are encoded in cross-platform quantum computer software developer kits and libraries. By testing the resulting programs on different quantum computer hardware architectures, one can then theoretically determine whether they will deliver consistent results.

As quantum hardware evolves, it becomes more and more difficult to prove that the results generated are correct. A solid foundation on which verification and explainability can be built into these systems is just as much part of the evolutionary process as the quantum hardware itself.

Facebook and Google took the hot seat on Tuesday during a hearing in the Senate subcommittee to discuss the Defend their companies’ business practices …

The FTC recently lifted its 2020 guidelines on vertical mergers. Meanwhile, Pres. Joe Biden plans to develop a new …

The technological advantages of 5G are well known. The hard part for business leaders is deciding where to implement the 5G …

Microsoft researchers believe Nobelium, the Russia-backed group that violated SolarWinds, is using a backdoor tool called …

An anonymous troubleshooter criticizing Apple’s handling of reports on its bounty program has details on three zero-day …

After cyberattacks like the Colonial Pipeline, US senators want a national law that mandates critical infrastructure companies …

Teams can check the network status before and after network changes to ensure a smooth deployment of the changes. Automating these pros …

Most organizations are slow to implement new techniques, including white box networking. Experts explain why and give advice on how …

Catos SASE Cloud is missing some functions and it may be necessary for customers to replace individual products in order to be able to fully use the platform. But …

To design a green, sustainable data center, consider ways to improve energy savings and efficiency. Then look into the green …

Instead of deciding between cloud or colocation, you should use both cloud and colocation. Colocation can support a hybrid cloud …

Learn the basics of computing units: what they are, what they do, who makes them and who takes them over. Then decide …

Get to know the advantages of cloud-based data management systems, common pitfalls and strategies when considering different data levels …

Learn more about the use of column-oriented databases and the large data model, Data warehouses and high performance queries …

Nutanix aims to simplify the management, provision and protection of data and databases in hybrid and multi-cloud environments …

All rights reserved,
Copyright 2000-2021, TechTarget
Privacy Policy

Cookie settings

Don’t sell my personal information

Ref: https://www.computerweekly.com