đź”’ What the heck are quantum computers and why do they matter? – The Wall Street Journal

by

(The Wall Street Journal) Imagine a world without traffic jams where computers use data to reroute cars and dissolve potential tie-ups before they even happen.

It might be closer than we think.
___STEADY_PAYWALL___

Quantum computers are a revolutionary way of computing that relies on the very strange laws of quantum physics to create machines so powerful, they could instantly crack encryption, unravel digital currencies, such as Bitcoin and revolutionize whole industries from aviation to pharmaceuticals.

For decades quantum computers seemed the stuff of science fiction. But today some limited and highly specific machines exist and the races on to build the world’s first fully functional or universal quantum computer able to do things no other machine can.

DAVIDE VENTURELLI: “The first commercial quantum computer was available by D-Wave Systems. Google decided to create their own group, Microsoft expanded their quantum computing group, IBM also decided to create a near-term device.  Startup Regetti Computing got funded and they decided to create their own quantum computer. Now you have a lot of approaches.”

And now quantum supremacy appears to be just around the corner.

It’s the tipping point where one of these machines does something impossible for any other computer.

In part one of this two part series, we look at how experiments at some of the world’s best known companies in fields as far flung as health, financial services and automotive are bringing us closer to the quantum age.

This is the future of everything. A look ahead from the Wall Street Journal.

From the newsroom in New York, I’m Jennifer Strong and this is the future of everything.

Hi, I’m Sarah Castellanos I work for WSJ CIO Journal and I cover emerging technologies.

That includes quantum computing, which she describes this way:

SARAH CASTELLANOS: “When for example, my parents asked me what quantum computing is I tell them think about it as sort of the next generation computer.”

Today’s computers carry information in units called bits which are either zeros or ones. But quantum bits or qubits can represent and store information in both zeros and ones at the same time.

SARAH CASTELLANOS:  “I mean, this is a faster, more efficient, completely new way of computing. Yes, it harnesses the power of quantum physics, and they’ll say, well, what is quantum physics?”

Microsoft founder Bill Gates recently told the Wall Street Journal it’s the one part  the company he truly doesn’t understand. Saying “I know a lot of physics and a lot of math but the one place where they put up slides and it’s hieroglyphics it’s quantum.

SARAH CASTELLANOS: “As more people start talking about it and mainstream as more companies come out and talk about their experiments with quantum computing, it’ll be a little bit more accessible. But the science itself is very dense.”

It helps to start with a little thought experiment called Schrodinger’s cat.

The story goes something like this. You put a cat in a box with a bottle of poison. The poison bottle could be open with the cat dead and the poison could be closed with the cat alive. At the same time.

Crazy as it sounds, scientists have proven atoms can exist in two states at once. It’s called superposition. And it means a single particle can be in two places at the same time.

The poison bottle is linked to the cat’s outcome, symbolizing entanglement. That’s what allow scientists to compute across quantum bits or qubits and it’s the basis of what allows a quantum computer to calculate things much, much faster than today’s machines or so called classical computers.

It’s okay to be confused. Even Einstein struggled with this.

Tech giants across the globe are racing to try to build a so called universal quantum computer, one that’s fully functional and can be scaled to perform a wide range of computations.

My name is Dario Gil and I’m the Vice President of artificial intelligence and quantum computing of IBM.

IBM’s TJ Watson Research Centre is about an hour north of New York City.

DARIO GIL: “One of the beautiful things that we can leverage with a quantum machine is now we can have instead of two states we can have two to the n, where n is a number of qubits and not limited to zeros and ones.” 

Decades after physicist Richard Feynman introduced this idea of quantum computing, what have we learned?

For one, it’s really hard to control qubits. Qubits are fragile. If there’s any change in temperature, noise, frequency or even motion, their quantum state will probably collapse.

DARIO GIL: “When you build a quantum computer that basic kernel is this creation of these qubits and these qubits are not perfect. They have errors in them.”

IBM, Google and other companies are working on error correction algorithms, but those require additional qubits to check the work of the ones running the computation.

DARIO GIL: “We only have a window of time, which can be reasonably short – it may be only 90, 100 microseconds – with which we could perform useful computation. Now, we are working in our laboratories to extend the window and time in which we can perform those computations. And as we keep extending it, at some point we will get to this universal fault tolerant quantum computer.”

The longer a cubic can stay in that magical quantum state, the more time you have to do useful things with it. We head into the lab to see one of these machines.

DARIO GIL: “This is what a quantum computer looks like.”

At first it’s hard to tell what we’re looking at. A large white cylinder hangs above us called a cryostat. It’s connected to a maze of racks and wires and deep inside this cryostat sits IBM’s tiny chip.

DARIO GIL: “Very often we make that analogy that this is like the 1940s in computers and you would see these computers that fill entire rooms….”

But what we’re seeing here looks and sounds quite different.

DARIO GIL: “One characteristic of these quantum computers is that they’re really beautiful. They’re very, very different right? We’re used to these square boxes and the chirping sound that you’re hearing is their compressor working to be able to flow this mixture of helium through the refrigerator to be able to cool it. So at the very bottom of it we get towards 15-milli kelvin, so very close to absolute zero. It’s 100 times colder than outer space at the bottom of it….”

This system requires extreme cold to work.

“My name is Jerry Chow, and we are in our quantum computing research lab here in aisle five of TD Watson research lab.”

He manages IBM’s experimental efforts with quantum computing.

JERRY CHOW: “In this lab is where we do a lot of the work in order to develop better and better quantum processors. This is where we explore things like materials and the way that we characterize the system so that we can make improvements to the way that we control our devices so they’re ready for building these quantum processes which end up going online or going to clients for people to actually make use of for science. They kind of look like large beer kegs and you certainly hear the noise which corresponds to these cryo compressors that keep them running and keep them cold.”

He has one of these cryostats open for us to look at inside there’s an organized tangle of tubes and wires in gold copper and silver. It’s all very steam-pump….

JERRY CHOW: “It goes inside of a bunch of shields which allow us to protect the device from radiation and other additional noise. There’s also a whole suite of electronics and components that sit outside of the refrigerator and these are used to control the qubits that we have in our quantum processor.”

Then we get a closer look at actual qubits under a microscope.

JERRY CHOW: “This printed circuit board has a number of inputs and outputs that allow you to drive the chip with microwaves. What you see is essentially the piece of silicon with various superconducting metallic traces that are on it, and those metallic traces define circuits and some of those circuits in fact contain what are known as Josephson junctions which form the basis of our quantum bit or Qubit.”

IBM recently launched the Q-network. Companies work with researchers to experiment with quantum computing use cases that makes sense for their businesses.

My name is Bob Stolte and I’m the Chief Technology Officer for equities at JP Morgan’s corporate investment bank.

JPMorgan is looking for ways that technology could be useful to the financial services industry.

BOB STOLTE: “Financial modelling pricing of our instruments, running risk scenarios, valuation arrest scenarios – there’s a lot of things that we do every day in the investment bank and in JPMorgan more broadly, all of which if and when the technology develops to a point that it surpasses what you can do with an existing computer, all of which could be much quicker. And it could allow you to do more risk scenarios, it could allow you to model instruments in ways that you can’t today.”

That’s because modern trading and portfolio management require a great deal of computing power.

BOB STOLTE: “So one way in which you try to determine how something will behave in the future, is you look at lots of ways that things have behaved in the past. And so you run countless scenarios of: if volumes were higher or lower; if prices were higher or lower; if the price of a stock or a trading pattern of something were to behave one way or another way….you need to process all that information through these algorithmic models to come up with all these different scenarios and probabilities associated with them. It takes an enormous amount of computing power. We’ve moved from what seemed very, very theoretical to what was in some sense a prototype, almost a toy. But the reason that people are starting to get excited and engaged is because we are on the doorstep of that potentially changing.”

MARTIN HOFMANN: “Sometimes people think it’s crazy what we do…..”

That’s Volkswagen CIO Martin Hoffman in Berlin. He’s referring to VW’s experiments with quantum computers.

MARTIN HOFMANN: “The first one that we worked with is a company called D-Wave in Canada and their approach to quantum computing is called quantum annealing.”

This approach is useful for solving optimization and logistics problems.

MARTIN HOFMANN: “One thing that comes to mind immediately is traffic optimization because we all know what it is to be in a traffic jam. The question is, can we use a quantum processor with infinite compute power to dissolve traffic jams before they even happen.”

VW ran this experiment with data from one of the world’s most congested cities -Beijing.

MARTIN HOFMANN: “So we used available data from the city of Beijing and basically what we did was first step, predicting, based on the data, when traffic jams were going to happen and then move that over to the quantum computer to let it dissolve the traffic jams by optimizing every single car in that traffic cloud. And by doing that we could demonstrate that problem could be applied successfully to a quantum computer using the D-Wave machine.”

It calculated the fastest routes to the airport while completely minimizing traffic congestion. Other computers would have taken 45 minutes to complete this task, but for a quantum computer it took….

MARTIN HOFMANN: “…..under a second.”

Though he says it’ll still be a while before we see this applied in real life.

MARTIN HOFMANN: “It doesn’t mean that you can prevent any traffic jam in the future in the world, but at least now we have an inroad when we know what kind of technological approach we could use to make traffic much more manageable.”

Hoffman thinks quantum computing could also improve the batteries in electric cars. For that, VW is partnering with Google.

MARTIN HOFMANN: “The idea is to develop an algorithm that allows to simulate the battery behaviour because it’s very close to quantum physics.”

He says there’s presently no simulation software for these batteries. So you have to build prototypes with quantum computing. Hoffman believes he could simulate their behaviour much faster and that could lead to reduced charging time VW is also working with Google to explore how AI and quantum computing could work in tandem.

MARTIN HOFMANN: “If you can combine quantum computing and AI, that’s a very powerful combination because you get incredible compute power with the capabilities we all know about AI. And self-driving cars requires a lot of algorithmic development. So we could use that approach to help develop algorithms for self-driving cars in a different way.”

And he says companies need to get on top of these developments. Now.

MARTIN HOFMANN: “I mean if you just do the math and follow the idea of exponential growth in technology, it’s going to be harder and harder to catch up. Technology is not an excuse anymore. You can do anything with technology. Technology is not a limiting factor.”

My name is Davide Venturelli, I’m science operations manager at USRA of the Research Institute for Advanced Computer Science. We run the program at NASA on the Quantum Artificial Intelligence Laboratory.

He believes NASA helped spark the now widespread industry interest in quantum computing.

DAVIDE VENTURELLI: “Our group was the first to believe in potential near-term applications and to give also some trust to the first commercial quantum computer which was available by D-Wave Systems.”

NASA held a conference in 2012 called quantum future technologies.

DAVIDE VENTURELLI: “And that conference was to seed for the creation of a research group and then for Google to decide to back financially the project, and this really gained the attention worldwide. And shortly after our group was founded Google decided to create their own group, Microsoft expanded their quantum computing group, IBM also decided to create a near-term device.  The startup Regetti Computing got funded and they decided to create their own quantum computer. Now you have a lot of approaches. We work with Regetti. We work with Google. We work with IBM. We work with D-Wave. And we have access to their prototypes and we experiment with them heavily by suggesting modifications, improvements and by most importantly, try to use them for something somewhat useful. So problems which are in optimization, problems which are the design of best routes, the design of best schedules, the design of best circuitry. Drug discovery for sure. It’s also in other aspects, for example catalysers or fertilizers. So there are certain chemical compounds that could have a tremendous impact even climate change or in how we efficientize agriculture.”

The question is, can we do it.

DAVID VENTURELLI: “It’s not an engineering challenge. It’s not like how to go to Mars. We know we can go to Mars. We went to the moon. We know it’s possible. Okay. But we’re not 100% sure that quantum computing will allow us to solve the problems we want as fast as we want. And we don’t know if that’s possible, even in principle. What we will learn in this journey is much more than having a computer which solves a problem faster, which is nice. But we will really learn how the world works, and potentially even some why’s of why the world works.”

SARAH CASTELLANOS: “The entire industry is still nascent, but a few executives from these companies have spoken about the importance of investing in this emerging technology now so that their employees and their customers will be totally ahead of the curve when quantum computers do become scalable and widely available.”

One of these is the biotech company Biogen

SARAH CASTELLANOS: “They completed an experiment last summer that showed that quantum computers have the potential to speed up drug discovery for diseases like multiple sclerosis, Alzheimer’s and Parkinson’s. So through a partnership with Accenture and 1Qbit, Biogen tested how quantum computers could help speed up the process of molecular matching. This is an important step in early phase drug design and discovering. Because basically the structure of molecules and their chemical features can be used to predict positive and negative effects of specific drugs on the human body. Like for example, the toxicity of a molecule.”

Molecular matching is a process that takes classical computers a really long time with many steps.

SARAH CASTELLANOS: “So it’s been described to me as it’s kind of like comparing a triangle and a square where the computer would start by comparing one corner and then an edge and then another corner and so on.

Rotating the molecules and seeing how they match up. So quantum computers can essentially be aware of all of the different overlaps simultaneously. It takes a very long time for drugs to be discovered, and then go from discovering to trial and eventually to, you know, the pharmacy and if there’s any chance that we can use a computer or a new type of computing method to help speed the process along that could be absolutely revolutionary.”

Visited 59 times, 1 visit(s) today

BizNews