Comments about the NEWS article in Nature: Quantum cloud goes commercial

Following is a discussion about this NEWS article in Nature Vol 543 9 March 2017, by Davide Castelvecchi
In the last paragraph I explain my own opinion.


The article starts with the following sentence:
IBM Q will not outperform conventional computers, at least not yet.
I have great doubt, based on my personnal 'experience' if it will ever do.
Quantum computers harness the counterintuitive properties of subatomic physics in which bits of information - called quantum bits or qubits - can assume multiple states simultaneously, rather than etc.
Nothing about qubits is counterintuitive. The issue is to demonstrate by means of experiments what it means that 'bits of information' can be simultaneously in multiple states. What is even more important is to demonstrate that you also can perform logical operations on these multiple states. The issue is here physics not mathematics.
Starting in 1990, theoretical physicists, have developed qubit based algorithms that in theory could perform certain tasks exponentially faster than classical computers can.
It is rather easy to claim how something should perform "In theory". The problem is to build something "In practice".

"Quantum Competition"

But in practice, getting enough qubits to wotk together to run any such algorithm - in what is known as a universal quantum computer - has proved extremely challenging.
The first is issue with a Quantum Computer is foremost that the larger the problem the more hardware is required. The same problem exist with an analog computer. Digital computers do not have this problem.
The second issue is when you have a problem to define a specific algorithm which works on a Quantum Computer. In general most problems can be solved on a digital computer. To define an algorithm which works on a QC is much more challenging. To define a solution which works faster on a QC than on a DC IMO can only be performed for a small set of specific problems and maybe the size of this set is close to zero.
In recent years, Google has also entered the fray (arena), establishing a superconducting qubit lab in Santa Barbara, California.
Let us wait and see.
But these machines would need to run on roughly 50 qubits each. The current record is about 20 qubits, barely enough for simple computations.
It will be interesting to see a list with shows the type of problems they can solve.
Even though it is not a state of art machine, IBM had to overcome several chalenges to get Quantum Experience online and make it usable for researchers who are not necessarily physicists and have never worked on a quantum computer before.
The most important advantage of Quantum Experience is that you get some hands on feeling what a Quantum Computer is. It is very easy to configure an experiment by combining quantum tools from a tool box, but it is very difficult to build an experiment that does what you have in mind. The most important task is to develop an algorithm which actual does that, without having the certainty that such an algorithm exists.

Era of the Quantum Cloud

"The real challenge is whether you can make your algorithm work on real hardware that has imperfections" says Isaac Chuang
The problem is that the Quantum Computer is based on probabilities. That means even if the QC works perfect you do not always get the same result. (A DC always gives the same result). That means most results are correct, but some are wrong. If the QC does not work perfect more results are wrong (without a warning). Of course this is a serious problem.
Quantum Experience has so far attracted about 40000 users from more than 100 countries. They have performed 275000 experiments and produced about 15 research papers.
See Reflection - Quantum Experience.
in which they compared the performance of IBM's superconducting machine with that of a five-qubit ion trap system at Monroe's lab (N.M. Linke et al Preprint at; 2017)
See Reflection - Quantum Experience.
The company's system was faster, but Monroe's system was more precise.
In fact both system's should be 100% precise.

Reflection 1 - Quantum Experience.

To participate select:
Quantum Experience 2.0 consists of 5 parts:
1) Community 2) User Guide 3) Composer 4) QASM Editor 5) My Scores
  1. Community. The Community consists of 4 parts: 1)Forum 2)Events 3)News and 4)Videos.
    • As of 17 April 2017 the Forum consists of 302 postings or subjects.
      My own interpretation is that level of the subjects discussed is not very high.
  2. User Guide. The User Guide consist of two guides: Beginners Guide and Full Users Guide
    My interpretation is that when you start with the Beginners Guide it seems all rather easy and IBM has done an impressive job, but when you want to fully understand the Full Users Guide it is not that easy.
  3. Composer. The Composer is used to build your own experiment based on 5 Qubits. There are two ways to perform an experiment: 1) Using a Real Quantum Computer and 2) Using Custom Topology.
    In the first case running your experiment costs some virtual money. In the second case the experiment is simulated on a Digital Computer and is for free. The final results are evaluated by performing each experiment 100 times.
  4. QASM Editor
  5. My Scores. This section services as a personnal database and contains the results of all your experiments. This is all very professional.

Reflection 2 - Hadamard and Grover Experiment

Reflection 3 - Perform a+b sum

The following Quantum Experience document explains how you implement an addition in a Quantum Computer:
For additional information select this link:

Reflection 4 - Shor's Algorithm
"Factoring with Qutrits: Shor’s Algorithm on Ternary and Metaplectic Quantum Architectures" by Alex Bocharov, Martin Roetteler ea.

If you want to give a comment you can use the following form Comment form

Created: 15 April 2017

Back to my home page Index
Back to Nature comments Nature Index