Andrew Fursman: Digital annealer — A stepping-stone to quantum computing


My name’s Andrew Fursman. I’m the CEO of 1QBit. There are fundamentally different paths that these companies that are producing quantum hardware are going down. The two biggest distinctions that you see are between what’s known as the circuit model of quantum computing, where we build the kind of gates that we see in classical computers, and something that’s more of a natural computing paradigm, known as quantum annealing where you basically think about a system that’s more like a natural system that evolves into a low energy state that answers these optimization problems. And so what we’ve seen is that there have been a pretty sharp divide between people who are pursuing the annealing path and people who are pursuing this more universal computing path. And in the early stages it was easier to build these quantum annealers and so that’s why you often hear people talking about optimization problems as the sweet spot for quantum computers and how we can harness the natural processes of the universe in order to answer these optimization problems. The quantum annealing paradigm is really exciting because even though it’s still very challenging to understand and transition to from classical computing it does really kind of have this native optimization framework. Digital annealers are a great first step in this direction but the problems that the digital annealers and the quantum annealers both answer are problems where it’s very easy to see why there is industrial relevance for this. Any time that you’re trying to say how can I get the best possible outcome while giving the least number of resources, that starts to look like almost every business that exists, right? I mean we always want to be finding the optimal way to be able to deliver these things. And so what I love so much about the Fujitsu Digital Annealer is that it’s something that makes use of everything that we already know about how to build classical computing systems and provides a device that solves the same problems as a quantum annealer. So it will probably scale much more like a classical computer scales but it’s something that will allow people to get their hands on these devices and understand how do you craft a problem for a device that’s very different in terms of its inputs and outputs than the types of machines that I’m typically used to dealing with. But we know that if you can build these specialized ASICs to solve specific problems then you can get incredible performance benefits. If you can, at the same time that you’re getting these performance benefits, also be learning how to compute in a model that should be accelerated by a completely new paradigm of computing, then that’s a huge bonus. Because you’re solving your problem today, you’re taking advantage of the specialization of an ASIC and you’re learning about a mode of computing that should be accelerated by this incredible explosion of funding and research into quantum devices. And so it’s just a win, win, win.

Leave a Reply

Your email address will not be published. Required fields are marked *