What’s up next for quantum computing? Possibly weather forecasting and online dating.
Dan Patterson, a Senior Producer for CBS News and CNET, interviewed futurist Isaac Arthur about what’s next for quantum computing. The following is an edited transcript of the interview.
Isaac Arthur: It’s always hard to guess with computers, and we’re a little bit spoiled by Moore’s Law from the fifties and sixties just taking us from these really simple devices to what we have nowadays.
We do not want to make the same mistake we made with, for instance, nuclear fission and fusion where we got the development in 20 years and just assume the next one will get to us in another 20.
Quantum computing might be many decades before we see any real major progress, but at the moment, we have made quite a few major leaps and actually are doing some real calculations with this.
SEE: Managing AI and ML in the enterprise 2020 (free PDF) (TechRepublic)
We have a whole bunch of problems in terms of making it better, though. The biggest one is actually getting the right answer out of it. As an example, if we were using the random source before–let’s say I locked somebody inside a quantum box with a phone book, and I told them, ‘I want you to find a phone number, and if you call this correct phone number and here’s the phone number in this book, someone’s going to come by and let you out of this box.’
That person is then given a random number generator, and we shut the box, and they search. A whole bunch of different quantum ghosts of them appear, searching various pages, but the one who finds the right one calls that, and the person comes and opens the door. That’s one example of a data extraction, though that would never work in actual reality because quantum doesn’t do both on the macroscopic scale, but you can get errors from things like that.
First, imagine one of these quantum people searching that page didn’t call the right number, but instead accidentally called a pizza delivery place that showed up and opened the door to deliver a pizza. Now, we have a wrong answer. We have things like this happen with quantum computing where we have an error, in terms of the data. We used to have this with normal computing too, but we solved it fairly early on.
This is probably going to be a lot harder to do, and in many ways, it’s the hardest part other than actually keeping all of these protocols entangled. It’s not just trying to keep one particle like this. We have to keep several thousand potentially–or millions–all entangled with each other simultaneously. This also allows them to be at just a hair above absolute zero temperature-wise. And then, of course, we have our third problem that has to be overcome, which is the software.
SEE: Augmented reality for business: Cheat sheet (free PDF) (TechRepublic)
All this runs on algorithms being had on class computers fed into these things, and those algorithms are the only way that we still have to do a lot of work on to improve them because we’re not quite using the original pure algorithms like Shor’s (algorithm), but ones we’ve had to adapt along the way. Those are kind of three areas–the software and the hardware areas are the ones that are going to really control limitations on advancing.
How much bigger can we make the entangled system? How well can we actually pull the right answer, and how do we actually get the right algorithms to ask the right question, as well?
What we tend to think–you know, with the modern phone and the laptop–that this would be something you have at your home, that you’d have a quantum computer, but in fact you probably never actually have a quantum computer in someone’s house. They have to be run at such very low temperatures. Even though they are very small devices in terms of the entanglement, there’s so much associated equipment that isn’t likely to get too miniaturized. Most likely, you would always have class computers, and people access it through the cloud, and you’d just buy time–or get time–on a quantum computer that you will link up to.
The thing that we’re most likely–for one individual person to use, would probably be something like encryption, but for stuff that we would actually get to see on our computer would probably be stuff like weather forecasting, for instance. It has a lot of options to allow us to do way better weather forecasting than we do now.
There are a lot of other examples in terms of the science; there are great things. It might finally let us model how the lifestyle of abiogenesis in the deep oceans, which is one of those examples where our models can’t really be. We have approximation algorithms that we use to cover these really huge numbers, but they don’t really seem to be up to snuff for covering things like those chemical interactions in the early deep oceans, and then those same algorithms, ironically enough, would be the kind of things we’d use for dating services in terms of finding the most optimal match for a person based on not just a simplified number of traits.
We have to simplify traits, normally. Here, we could actually have a thousand different traits with a thousand different subtypes, and a quantum computer could actually match up and optimize all of those. And then of course, there’s the possibility of using election modeling.
Watch more interviews with Dan Patterson and Isaac Arthur
This is a syndicated post. Read the original post at Source link .