Hi everyone!
The other day, I visited D-Wave Systems inward Vancouver, Canada. It’s a companionship that makes cutting-edge quantum computers.
I got to larn a lot virtually quantum computers there, together with thus I’d similar to part roughly of what I learned at that topographic point amongst you lot inward this article.
The destination of this article is to give you lot an accurate intuition of what a quantum reckoner is using a uncomplicated example.
This article volition non require you lot to convey prior cognition of either quantum physics or reckoner scientific discipline to live on able to empathize it.
Okay, let’s acquire started.
What is a quantum computer?
Here is a one-sentence summary of what a quantum reckoner is:
A quantum reckoner is a type of reckoner that uses quantum mechanics together with thus that it tin perform sure enough kinds of computation to a greater extent than efficiently than a regular reckoner can.
There is a lot to unpack inward this sentence, together with thus permit me walk you lot through what it is precisely using a uncomplicated example.
To explicate what a quantum reckoner is, I’ll bespeak to start explicate a piddling flake virtually regular (non-quantum) computers.
How a regular reckoner stores information
Now, a regular reckoner stores data inward a serial of 0’s together with 1’s.
Different kinds of information, such every bit numbers, text, together with images tin live on represented this way.
Each unit of measurement inward this serial of 0’s together with 1’s is called a bit. So, a flake tin live on prepare to either 0 or 1.
Now, what virtually quantum computers?
A quantum reckoner does non utilization bits to shop information. Instead, it uses something called qubits.
Each qubit tin non solely live on prepare to 1 or 0, but it tin also live on prepare to 1 together with 0. But what does that hateful exactly?
Let me explicate this amongst a uncomplicated example. This is going to live on a somewhat artificial example. But it’s even together with thus going to live on helpful inward agreement how quantum computers work.
A uncomplicated illustration for agreement how quantum computers work
Now, suppose you’re running a go agency, together with you lot bespeak to motility a grouping of people from 1 place to another.
To decease on this simple, let’s tell that you lot bespeak to motility solely 3 people for now — Alice, Becky, together with Chris.
And suppose that you lot convey booked 2 taxis for this purpose, together with you lot desire to figure out who gets into which taxi.
Also, suppose hither that you’re given data virtually who’s friends amongst who, together with who’s enemies amongst who.
Here, let’s tell that:
Alice together with Becky are friends
Alice together with Chris are enemies
Becky together with Chris are enemies
And suppose that your destination hither is to split upward this grouping of 3 people into the 2 taxis to hand the next 2 objectives:
Maximize the number of friend pairs that part the same car
Minimize the number of enemy pairs that part the same car
Okay, together with thus this is the basic premise of this problem. Let’s start retrieve virtually how nosotros would solve this employment using a regular computer.
Solving this employment amongst a regular computer
To solve this employment amongst a regular, non-quantum computer, you’ll bespeak start to figure out how to shop the relevant data amongst bits.
Let’s label the 2 taxis Taxi #1 together with Taxi #0.
Then, you lot tin stand upward for who gets into which automobile amongst 3 bits.
For example, nosotros tin prepare the 3 bits to 0, 0, together with 1 to represent:
Alice gets into Taxi #0
Becky gets into Taxi #0
Chris gets into Taxi #1
Since at that topographic point are 2 choices for each person, at that topographic point are 2*2*2 = 8 ways to split upward this grouping of people into 2 cars.
Here’s a listing of all possible configurations:
A | B | C
0 | 0 | 0
0 | 0 | 1
0 | 1 | 0
0 | 1 | 1
1 | 0 | 0
1 | 0 | 1
1 | 1 | 0
1 | 1 | 1
Using 3 bits, you lot tin stand upward for whatsoever 1 of these combinations.
Computing the grade for each configuration
Now, using a regular computer, how would nosotros create upward one's heed which configuration is the best solution?
To do this, let’s define how nosotros tin compute the grade for each configuration. This grade volition stand upward for the extent to which each solution achieves the 2 objectives I mentioned earlier:
Maximize the number of friend pairs that part the same car
Minimize the number of enemy pairs that part the same car
Let’s merely define our grade every bit follows:
(the grade of a given configuration) = (# friend pairs sharing the same car) - (# enemy pairs sharing the same car)
For example, suppose that Alice, Becky, together with Chris all acquire into Taxi #1. With 3 bits, this tin live on expressed every bit 111.
In this case, at that topographic point is solely 1 friend span sharing the same car — Alice together with Becky.
However, at that topographic point are 2 enemy pairs sharing the same car — Alice together with Chris, together with Becky together with Chris.
So, the full grade of this configuration is 1-2 = -1.
Solving the problem
With all of this setup, nosotros tin in conclusion decease virtually solving this problem.
With a regular computer, to honor the best configuration, you’ll bespeak to essentially decease through all configurations to come across which 1 achieves the highest score.
So, you lot tin retrieve virtually constructing a tabular array similar this:
A | B | C | Score
0 | 0 | 0 | -1
0 | 0 | 1 | 1 <- best="" font="" of="" one="" solutions="" the="">
0 | 1 | 0 | -1
0 | 1 | 1 | -1
1 | 0 | 0 | -1
1 | 0 | 1 | -1
1 | 1 | 0 | 1 <- best="" font="" other="" solution="" the="">
1 | 1 | 1 | -1
As you lot tin see, at that topographic point are 2 right solutions here — 001 together with 110, both achieving the grade of 1.
This employment is fairly simple. It speedily becomes likewise hard to solve amongst a regular reckoner every bit nosotros increase the number of people inward this problem.
We saw that amongst 3 people, nosotros bespeak to decease through 8 possible configurations.
What if at that topographic point are 4 people? In that case, we’ll bespeak to decease through 2*2*2*2 = sixteen configurations.
With n people, we’ll bespeak to decease through (2 to the ability of n) configurations to honor the best solution.
So, if at that topographic point are 100 people, we’ll bespeak to decease through:
2¹⁰⁰ = 10³⁰ = 1 meg million meg million meg configurations.
This is merely impossible to solve amongst a regular computer.
Solving this employment amongst a quantum computer
How would nosotros decease virtually solving this employment amongst a quantum computer?
To retrieve virtually that, let’s decease dorsum to the instance of dividing 3 people into 2 taxis.
As nosotros saw earlier, at that topographic point were 8 possible solutions to this problem:
A | B | C
0 | 0 | 0
0 | 0 | 1
0 | 1 | 0
0 | 1 | 1
1 | 0 | 0
1 | 0 | 1
1 | 1 | 0
1 | 1 | 1
With a regular computer, using 3 bits, nosotros were able to stand upward for solely 1 of these solutions at a time — for example, 001.
However, amongst a quantum computer, using 3 qubits, nosotros tin stand upward for all 8 of these solutions at the same time.
There are debates every bit to what it agency exactly, but here’s the way I retrieve virtually it.
First, examine the start qubit out of these 3 qubits. When you lot prepare it to both 0 together with 1, it’s sort of similar creating 2 parallel worlds. (Yes, it’s strange, but just follow along here.)
In 1 of those parallel worlds, the qubit is prepare to 0. In the other one, it’s prepare to 1.
Now, what if you lot prepare the bit qubit to 0 together with 1, too? Then, it’s sort of similar creating 4 parallel worlds.
In the start world, the 2 qubits are prepare to 00. In the bit one, they are 01. In the 3rd one, they are 10. In the quaternary one, they are 11.
Similarly, if you lot prepare all 3 qubits to both 0 together with 1, you’d live on creating 8 parallel worlds — 000, 001, 010, 011, 100, 101, 110, together with 111.
This is a foreign way to think, but it is 1 of the right ways to translate how the qubits comport inward the existent world.
Now, when you lot apply roughly sort of computation on these 3 qubits, you lot are genuinely applying the same computation inward all of those 8 parallel worlds at the same time.
So, instead of going through each of those potential solutions sequentially, nosotros tin compute the scores of all solutions at the same time.
With this item example, inward theory, your quantum reckoner would live on able to honor 1 of the best solutions inward a few milliseconds. Again, that’s 001 or 110 every bit nosotros saw earlier:
A | B | C | Score
0 | 0 | 0 | -1
0 | 0 | 1 | 1 <- best="" font="" of="" one="" solutions="" the="">
0 | 1 | 0 | -1
0 | 1 | 1 | -1
1 | 0 | 0 | -1
1 | 0 | 1 | -1
1 | 1 | 0 | 1 <- best="" font="" other="" solution="" the="">
1 | 1 | 1 | -1
In reality, to solve this problem, you lot would bespeak to give your quantum reckoner 2 things:
All potential solutions represented amongst qubits
A component that turns each potential solution into a score. In this case, this is the component that counts the numbers of friend pairs together with enemy pairs sharing the same car.
Given these 2 things, your quantum reckoner volition spit out 1 of the best solutions inward a few milliseconds. In this case, that’s 001 or 110 amongst a grade of 1.
Now, inward theory, a quantum reckoner is able to honor 1 of the best solutions every fourth dimension it runs.
However, inward reality, at that topographic point are errors when running a quantum computer. So, instead of finding the best solution, it powerfulness honor the second-best solution, the 3rd best solution, together with and thus on.
These errors acquire to a greater extent than prominent every bit the employment becomes to a greater extent than together with to a greater extent than complex.
So, inward practice, you lot volition likely desire to run the same functioning on a quantum reckoner dozens of times or hundreds of times. Then pick the best effect out of the many results you lot get.
How a quantum reckoner scales
Even amongst the errors I mentioned, the quantum reckoner does non convey the same scaling number a regular reckoner suffers from.
When at that topographic point are 3 people nosotros bespeak to split upward into 2 cars, the number of operations nosotros bespeak to perform on a quantum reckoner is 1. This is because a quantum reckoner computes the grade of all configurations at the same time.
When at that topographic point are 4 people, the number of operations is even together with thus 1.
When at that topographic point are 100 people, the number of operations is even together with thus 1. With a unmarried operation, a quantum reckoner computes the scores of all 2¹⁰⁰ = 10³⁰ = 1 meg million meg million meg configurations at the same time.
As I mentioned earlier, inward practice, it’s likely best to run your quantum reckoner dozens of times or hundreds of times together with pick the best effect out of the many results you lot get.
However, it’s even together with thus much improve than running the same employment on a regular reckoner together with having to repeat the same type of computation 1 meg million meg million meg times.
Wrapping up
Special thank you lot to everyone at D-Wave Systems for patiently explaining all of this to me.
D-Wave of late launched a cloud surroundings for interacting amongst a quantum computer.
If you’re a developer together with would similar genuinely to endeavor using a quantum computer, it’s likely the easiest way to do so.
It’s called Leap, together with it’s at https://cloud.dwavesys.com/leap. You tin utilization it for gratis to solve thousands of problems, together with they also convey easy-to-follow tutorials on getting started amongst quantum computers 1 time you lot sign up.
Footnote:
In this article, I used the term “regular computer” to refer to a non-quantum computer. However, inward the quantum computing industry, non-quantum computers are commonly referred to every bit classical computers.
Buat lebih berguna, kongsi:
