(Image source: Internet) Analog computers have been providing services to mankind long before digital computers swept the world. Now, the nature of quantum physics is causing us to reexamine the distinction between digital and analog. Quantum hardware companies like Pasqal are “e

(Picture source: Internet)

Long before digital computers swept the world, analog computers had already provided services to mankind. Now, the nature of quantum physics is making us reexamine the distinction between digital and analog. Quantum hardware companies like Pasqal are "embracing" this duality by developing hybrid digital/analog solutions to accelerate the realization of industrial quantum advantage.

Digital and Analog Computing

The wonderful thing about computers is that it can turn questions into answers. Humans have been using computers for thousands of years: as early as the 2nd century BC, the ancient Greeks used manual computers (Antikythera calculating machines) to predict astronomical positions and solar eclipses.

So, what makes today’s computers special? Why do quantum computing and need to take into account this duality?

First of all, it takes a certain amount of time for computers to calculate problems. The entire process includes continuous and discrete types, which leads to the distinction between analog and digital. Currently, digital computing is the dominant method and has even become synonymous with computing. Numerical computing dominates for several reasons: versatility and fault tolerance, among others.

Before 1960, the two calculation methods were on equal footing.

Analog computing brought computing advantages to humans earlier than digital computers. For example, in 1836, the French mathematician Gaspard-Gustave Coriolis introduced an analog computer for solving differential equations . These computers continued to develop during the 19th and early 20th centuries and began to be used to predict tides and projectile trajectories. From 1960 to 1984, the Delta analog computer (Figure 1) was used to design and implement the Delta Works large-scale infrastructure project, which was designed to protect densely populated areas of the Netherlands from maritime threats.

DeltarAnalog computers

Revisiting analog computing with quantum

Quantum computers are also divided into digital and analog forms: In digital quantum computing, we perform algorithms by implementing sequences of discrete operations, often called quantum gates; in On the quantum computing simulator, users can control a small number of parameters to continuously get closer to the answer.

One of the main reasons why classical computing has gone digital is fault tolerance. This is because the "rounding" rule can eliminate small errors, such as 0.99 is treated as 1, 0.002 is treated as 0, bit is very robust to small errors. Furthermore, every possible calculation can be programmed using number operations, that is, digital computers are universal.

In classical computing, analog computers are used to solve specific tasks. Characteristics aside, analog computers can be used as powerful and energy-efficient accelerators.

Furthermore, the nature of analog and digital is even more subtle in the quantum realm. Continuity and discreteness are inextricably linked in quantum systems, such as wave-particle duality . qubits are analog objects until they are measured and turned into numbers. This means that quantum computers, even digital ones, can retain their analog properties. We believe that quantum computers should embrace this duality and exploit it as a resource.

Digital/Analog Duality

In those neutral-atom quantum processors with dual digital/analog capabilities, computation follows three main steps: First, the input information is encoded into an array of neutral atoms; then, the information is passed through the tuned laser for processing; then the final state is read out by taking a "picture" of atoms to extract the calculation results.Neutral atom computers can run processing steps in digital/analog mode or a combination of both modes, enabling quantum advantage in the near/long term.

The main advantage of the numerical method is versatility. Every possible operation on a quantum computer can be represented as a (finite) sequence of gates. The problem is that today's quantum computers can only execute extremely short sequences before the results become unreliable.

This is because a single gate is very noisy and error complex. Therefore, even large (multi-qubit) digital quantum computers are very limited in their capabilities. In comparison, the emulation mode is slightly less versatile but less susceptible to noise. So just like the development of classical computing, quantum simulation methods are expected to deliver early quantum advantages in specific but critical applications.

It should be emphasized here that this method is different from the more restrictive simulation method of quantum annealing.

Quantum annealing serves as a computing paradigm that can solve specific classes of problems, such as optimization problems. Although, in theory, any problem can be converted into an optimization problem, quantum annealing is better suited to tasks that already have this form. The simulation calculations of neutral atom quantum computers can perform far more tasks: in addition to optimization tasks, they include chemistry, physics and materials engineering. Furthermore, it is possible to combine digital gates (digital analog approach).

Simulated Quantum Advantage

The simulation mode of the neutral atom quantum computer embodies Richard Feynman 's dream of using synthetic quantum systems to simulate nature. Neutral atom quantum computers have now surpassed classical analog calculations in simulating physical systems: they have been used to simulate antiferromagnetism in two-dimensional materials , as well as for the implementation of two iconic quantum magnetism models Programmable simulator.

These achievements are based on the scalability and fault tolerance advantages of the neutral atom quantum processor in simulation mode: hundreds of qubits can be effectively operated in simulation mode and a variety of simulation operations can be achieved.

There are several neutral atom quantum hardware manufacturers around the world, such as Pasqal, QuEra Computing, ColdQuanta and Atom Computing. They are taking steps to bring quantum advantages to industrial applications, including Pasqal, which is studying simulation methods, Atom, which is studying gate methods, etc. .

One of the very promising directions is solving differential equations. To advance the program, Pasqal introduced a powerful differential equation solver compatible with near-term quantum hardware and tested it on paradigm engineering problems such as modeling structural integrity.

Interestingly, the performance of the solver is improved when executed in mixed digital/analog mode. While single-qubit gates are used to tune and guide output solutions, multiple qubits are manipulated to achieve the power of "entanglement."

In digital mode, these entanglement operations are broken down into a large number of two-qubit gates. However, in a simulated environment, the natural interaction forces between atoms can effectively entangle them together in a short period of time, with errors reduced, and the entire system moving efficiently towards the desired outcome.

Differential equations play a vital application role in a large number of industries. Therefore, the impact of quantum advantage in this field is huge.

Neutral Atom Quantum Computer is relying on simulation mode to implement quantum enhanced machine learning technology. So far, their focus has been on graph-related problems, such as quantum computing company QuEra publishing promising results solving the maximum independent set problem, which has broad applications in industries including , telecommunications, , finance and logistics. .

Pasqal showed how to use simulated quantum methods in the Quantum evolution kernel method (Quantum evolution kernel, QEK) to solve graph classification problems, which has been proven to be superior to classical kernel methods in terms of accuracy and computational cost. Pasqal also presented the first application of QEK in compound toxicity screening, which they are investigating to determine optimal chemical reaction pathways.

Today, Pasqal can achieve entanglement and manipulation using hundreds of qubits in simulation mode and leverage this ability to solve multiple exciting problems.

More and more quantum hardware manufacturers are embracing the analog/digital dual characteristics of quantum computers. Taking a lesson from history, Pasqal believes that analog quantum computers are bringing quantum advantages.

Original link:

https://thequantuminsider.com/2022/06/28/why-analog-neutral-atoms-quantum-computing-is-a-promising-direction-for-early-quantum-advantage/

Text: Loïc Henriet (CTO, Pasqal)

Compiler: Li Mi

Editor: Mu Yi

Note: This article is compiled from "thequantuminsider" and does not represent the views of Quantum Outpost.

To learn more about quantum computing information, please follow the WeChat public account "Quantum Outpost"!