In a recent paper, we showed that secondary storage can extend the range of quantum circuits that can be practically simulated with classical algorithms. Here we refine those techniques and apply them to the simulation of Sycamore circuits with 53 and 54 qubits, with the entanglement pattern ABCDCDAB that has proven difficult to classically simulate with other approaches. Our analysis shows that on the Summit supercomputer at Oak Ridge National Laboratories, such circuits can be simulated with high fidelity to arbitrary depth in a matter of days, outputting all the amplitudes.
Metadata: arXiv. Abstracts retain their original authors’ rights.
In the conversation
Original HN comments, with labels from simple text rules. A label describes the whole comment and may miss context or irony. Read the discussion to make up your own mind. About the labels
My understanding is that this simulation of random quantum circuits is already known to be hard in a theoretical sense[1], and there is no argument that a classical computer could compete with a quantum computer much bigger than 50-60ish qubits. And that arguments against those previous papers has more to do with classical optimizations such as using disk w/ RAM rather than just RAM alone.
For example, take this line from IBM's refutation of Google's Quantum Supremacy experiment: "64 PiB of disk space are required for 53-qubit circuits, and 128 PiB for 54-qubit circuits. Both fit within the 250 PiB available on Summit."[2] If you have to use over 50% of the total disk…
Read the full comment
space on one of the world's strongest supercomputers, you're probably going to run into some major issues once we start hitting the 55, 56, 57, 58, etc. qubit territory. And although this is a moving goalpost, the supercomputers clearly aren't going to get better faster than adding single qubits onto the current chips. Already Rigetti is announcing an 80-qubit quantum computer, and IBM is making an 127-qubit computer.[3]
So, IMHO we are far past the point where quantum computers have to prove themselves superior to classical ones, at least on designed tasks. I don't think we should compare failed P vs NP proofs (which, IMHO is a problem we're anywhere from 5 to 10000 years out from solving).
All this being said, I do have to agree with IBM on the value of the term "Quantum Advantage" versus "Quantum Supremacy". Quantum computers have not yet demonstrated a significant advantage in terms of cost on any problem, but I'm optimistic they will in about a decade. Probably in problems that are already naturally fit to quantum computers, like studying interactions between particles in quantum physics. Also for the record I'm a layperson with just an unusual interest in this field, so it's easily possible I've missed some incredibly important central problem that everyone is being quiet about.
[1] https://arxiv.org/abs/2007.07872
[2] https://arxiv.org/pdf/1910.09534.pdf
[3] https://arstechnica.com/science/2021/06/quantum-computing-st...
The Quantum Computer scientists Scott Aaronson and Gil Kalai have casted doubt on this result[1][2].
Also good to note is that Kalai casted doubt on the original Google supremacy result. There are now classical algorithms 6 orders of magnitude faster than the Google quantum experiment[3].
1.https://www.scottaaronson.com/blog/?p=5122
2.https://gilkalai.wordpress.com/2020/12/06/photonic-huge-quan...
3.https://arxiv.org/abs/1910.09534
If a worst case to simulate on a classical computer is 2.5 days (per IBM's blog) then they should either:
1. Actually preform the experiment per their approach, compare the results, and provide tangible proof in the coming days.
or
2. Retract their claims.
In either case, exciting times ahead.
[Edit: Yes, I read their published paper[1] which only claims it is possible to achieve and does not perform the same experiment.
1. https://arxiv.org/pdf/1910.09534.pdf]
Fast compilation is important when compilation occurs at runtime, such as query compilers in modern database systems and WebAssembly virtual machines in modern browsers.
In honor of Alan Turing's hundredth birthday, I unwisely set out some thoughts about one of Turing's obsessions throughout his life, the question of physics and free will.