The world’s leading publication for data science, AI, and ML professionals.

Is Quantum Supremacy a Thing or Not?

Google says that they did it, quantum computers are not only real, they are currently demolishing classical computers, others disagree.

If you are immersed in the computer world you may have heard that Google reached something called the "Quantum Supremacy". What is that? This is what they said:

We developed a new 54-qubit processor, named "Sycamore", that is comprised of fast, high-fidelity quantum logic gates, in order to perform the benchmark testing. Our machine performed the target computation in 200 seconds, and from measurements in our experiment we determined that it would take the world’s fastest supercomputer 10,000 years to produce a similar output.

So the quantum computer solved the problem in 200 seconds (around 3.3 mins) and it would take a classical supercomputer 10,000 years to do the same! That’s astonishing. Then I watched this video:

I was amazed!! And I started writing this article to show how amazing the results were. And then! BUM!!! I saw this:

On "Quantum Supremacy" | IBM Research Blog

Let me paint you the whole story here. The Google paper, published today October 23 on Nature, was kinda leaked before in NASA’s NTRS webpage, but it got deleted with no explanation. Everyone was like: whaat the heellll??

Then IBM published that blog, together with this article:

Leveraging Secondary Storage to Simulate Deep 54-qubit Sycamore Circuits

Where they said:

We argue that an ideal simulation of the same task can be performed on a classical system in 2.5 days and with far greater fidelity. This is in fact a conservative, worst-case estimate, and we expect that with additional refinements the classical cost of the simulation can be further reduced.

So not 10,000 years? 2,5 days? That’s a big difference. To understand the technical parts of this discussion read this great blog by Scott Aaronson:

Shtetl-Optimized

Where he explains and gives his opinion about this problem. But there’s hope in Scott’s blog:

But does IBM’s analysis mean that "quantum supremacy" hasn’t been achieved? No, it doesn’t – at least, not under any definition of "quantum supremacy" that I’ve ever used. The Sycamore chip took about 3 minutes to generate the ~5 million samples that were needed to pass the "linear cross-entropy benchmark" – the statistical test that Google applies to the outputs of its device. Three minutes versus 2.5 days is still a quantum speedup by a factor of 1200.

So ok, we’re not in the point that we can say that quantum computers are 1752000 (I calculated that) faster (or better) than classical computers, but 1200 times it’s enough.

We are at new beginnings for the quantum computational era. And with that comes quantum machine leaning (scary writing those words after the controversy), and a new world of possibilities for AI and human beings.


If you want to understand this short blog read in this order:

Quantum Supremacy Using a Programmable Superconducting Processor

On "Quantum Supremacy" | IBM Research Blog

Leveraging Secondary Storage to Simulate Deep 54-qubit Sycamore Circuits

Quantum Supremacy Using a Programmable Superconducting Processor

Quantum supremacy using a programmable superconducting processor

Shtetl-Optimized


Thanks for reading and keep learning all the time 🙂


Related Articles