以下内容由机器翻译生成。如果您觉得可读性不好, 请阅读原文或 点击这里.
It seems, these days, like we seldom go very long between quantum computing breakthroughs. Yet, somehow, skeptics remain convinced the media exaggerates the promise of quantum supremacy. We probably are, to varying degrees, but that doesn’t mean it won’t happen.
If all of science were a neighborhood, the field of quantum computing would reside at the corner of “I Need Funding” avenue and “A Few Eurekas Away” boulevard. And that’s why some experts feel supremacy, the idea that quantum computers will eventually outperform classical ones, is a pipe dream. History says those people are wrong. But before we get into that, let’s examine the issue.
What is quantum supremacy?
Simply put, it’s that special moment in our future’s history where a quantum computer could honestly taunt a classical one with the phrase “anything you can do, I can do better,” were quantum computers capable of taunting (and classical ones of caring). In essence, the point at which a quantum system is capable of outperforming a classical one at the same tasks.
In order to simplify the convervation, we need to have a broad definition for the term. For quantum computers to be usable they can’t just be multi-million dollar systems dedicated to solving a single problem. That’s not a revolutionary new technology that could change our fundamental undestanding of the universe, as we’ve written in our previous coverage.
Instead, let’s say that quantum supremacy would be the creation of a quantum system capable of running a myriad of algorithms, for a large variety of different applications, that all classical (binary) systems are incapable of running.
How far away is it?
Unfortunately, nobody knows (here’s why). That doesn’t mean it won’t happen — nobody knew when humans would unlock the secrets of flight or develop the Large Hadron Collider. These things don’t happen in flashes like hurricanes that change the landscape, but over decades of agonizingly slow, underfunded, misunderstood research.
在最近发布的 文章, TNW contributor Colin Earl says:
Researchers have been predicting such a breakthrough “within the next decade” for almost forty years now and despite the billions of dollars poured into development, the target date keeps retreating.
I’d argue that the sheer volume of jaw-droppingly exciting breakthroughs in the field of quantum computing extend beyond my ability to cover them in a normal work week.
To be fair to Colin, he’s referring to a very specific problem: noise. He goes on to state:
Noise is not a crippling problem for traditional computers – their circuits can be made robust enough to stand some noise and whisk it away in the form of heat.
But the quantum state is intrinsically fragile and destroyed by noise. There are attempts to address this challenge with error correcting quantum circuits, but these mechanisms would also have to scale exponentially with the qubit count. They can’t. And so quantum computers will never be usable.
But is noise really the full stop – do not pass Go, do not collect $200 – problem that it’s being made out to be? Kinda. Sorta. Maybe.
Fixing the noise problem
Noise is arguably the biggest problem for quantum computer systems. In fact, I’d opine, overcoming that problem is basically your job if you’re developing these machines. You can read our quantum computing primer 这里, which has a section explaining key concepts such as observation theory and noise.
Luckily, just like when IBM and other hardware makers had to figure out how to shrink a mainframe into an iphone, we’re not on a deadline here — such noticeable change takes decades. Eventually, many scientists have asserted, we’ll figure it all out. And that’s not just conjecture. Here’s a list of potential solutions for the noise problem in quantum computers:
- Time crystals. Nobody understands these things, but they might be one answer
- Bose-Einstein condensate. This one has frozen atom clouds and lasers.
- A real-life flux capacitor. No kidding, it’s a quantum error-correction tool.
- New controls by Rigetti.
- 一个 experimental theory on creating more robust fault-tolerance in quantum systems.
- Further advancements in machine learning and AI.
And there’s dozens more in the form of ongoing experiments, maybe hundreds. As long as researchers are working on the noise problem, and the temperature one, and the scaling issues, and whatever other obstacles arise there’s hope.
We may not achieve quantum supremacy this year, as Google boasted, but that doesn’t mean we won’t in 20, 30, or 100 years. There’s not a lot of room for “never” statements in the world of quantum physics.