Getting Started with Quantum Computing in Python • Hacker News Board

Re: Getting Started with Quantum Computing in Python

#31

Post by throwaway37585 » Mon, Jul 23, 2018, 12:11 AM UTC

furqs wrote: post #10

I have a question. Please excuse my ignorance but I thought that once the world has a working quantum computer, the world as we know it will be destroyed. Since a quantum computer can solve any NP hard problem in a polynomial time, it would mean it could break any kind of crypto, any kind of security and can brute force anything. Why hasn't that happened yet since its 2018 and we already have quantum computers?

Also, keep in mind that NP hard ≠ NP complete. By saying 'solve any NP hard problem in polynomial time', you're also saying 'solve any NEXPTIME hard problem in polynomial time', which is known to be false.

Re: Getting Started with Quantum Computing in Python

#32

Post by wzeng » Mon, Jul 23, 2018, 12:39 AM UTC

Another open source Python framework is Forest from Rigetti Computing. https://www.rigetti.com/products

A simple getting started post on a similar topic is here: https://medium.com/rigetti/how-to-write-a-quantum-program-in... which might be useful for comparison.

Re: Getting Started with Quantum Computing in Python

#33

Post by peter_retief » Mon, Jul 23, 2018, 5:20 AM UTC

i cant think where i could possibly use this but it seems concise and easy to follow, so here i go..

Re: Getting Started with Quantum Computing in Python

#34

Post by seventhtiger » Mon, Jul 23, 2018, 5:51 AM UTC

sidcool wrote: post #4

But confused. How does this work without Quantum hardware?

Every turing machine can emulate every other turing machine. Different turing machines are all equivalent in what they can compute, but they can computer things at different speeds.

Both binary computers and quantum computers are turing machines.

Re: Getting Started with Quantum Computing in Python

#35

Post by peter_retief » Mon, Jul 23, 2018, 7:04 AM UTC

peter_retief wrote: post #33

i cant think where i could possibly use this but it seems concise and easy to follow, so here i go..

Ok, interesting, I dont really get the superposition on a logic gate though and can't see any relation to entanglement?

Re: Getting Started with Quantum Computing in Python

#36

Post by antpls » Mon, Jul 23, 2018, 10:55 AM UTC

throwaway37585 wrote: post #26

randomsearch wrote: post #24

Earlier quoted context omitted.

Great comment. Circuits are particularly unhelpful in describing quantum algorithms because they usually indicate a fixed problem size, and they do not provide an insight into entanglement, one of the fundamental “resources” of QC. We need new representations and much better abstractions to get away from the low-level thinking we are currently promoting.

> [Quantum circuits] do not provide an insight into entanglement What do you mean? Entanglement occurs whenever the state of a system cannot be factored into a product of the states of its components. Quantum circuits can definitely do that. Just take a qubit, apply the Hadamard gate to it, then CNOT it with a second qubit to get an entangled Bell state. You can see it in action here: http://demonstrations.wolfram.co…

I'm not sure about the actual intent of the grand-parent, but I think it was referring to insight that a larger audience can understand.

I have a CS degree, and I don't know what is "Hadamard", "Bell state", etc. Also your link doesn't load on Firefox Mobile Android, so it didn't help :-)

Proposing a Python library about Quantum Computing is an attempt at explaining mechanisms to a larger audience (surely it is not an easy task)

Edit : the link finally loaded on another tab while writing the comment, but you have to pay for a license of Wolf.A. to run it, I guess. That can't be arguably considered "accessible" knowledge.

Re: Getting Started with Quantum Computing in Python

#37

Post by randomsearch » Mon, Jul 23, 2018, 1:16 PM UTC

throwaway37585 wrote: post #26

randomsearch wrote: post #24

Earlier quoted context omitted.

Great comment. Circuits are particularly unhelpful in describing quantum algorithms because they usually indicate a fixed problem size, and they do not provide an insight into entanglement, one of the fundamental “resources” of QC. We need new representations and much better abstractions to get away from the low-level thinking we are currently promoting.

> [Quantum circuits] do not provide an insight into entanglement What do you mean? Entanglement occurs whenever the state of a system cannot be factored into a product of the states of its components. Quantum circuits can definitely do that. Just take a qubit, apply the Hadamard gate to it, then CNOT it with a second qubit to get an entangled Bell state. You can see it in action here: http://demonstrations.wolfram.co…

I mean, they do not illustrate when qubits have become entangled. You can't see that. I'm sure we can do better.

Re: Getting Started with Quantum Computing in Python

#38

Post by throwaway37585 » Mon, Jul 23, 2018, 3:33 PM UTC

randomsearch wrote: post #37

throwaway37585 wrote: post #26

Earlier quoted context omitted.

> [Quantum circuits] do not provide an insight into entanglement What do you mean? Entanglement occurs whenever the state of a system cannot be factored into a product of the states of its components. Quantum circuits can definitely do that. Just take a qubit, apply the Hadamard gate to it, then CNOT it with a second qubit to get an entangled Bell state. You can see it in action here: http://demonstrations.wolfram.co…

I mean, they do not illustrate when qubits have become entangled. You can't see that. I'm sure we can do better.

I’m not sure what you mean by “illustrate when qubits have become entangled”. Can you explain?

Re: Getting Started with Quantum Computing in Python

#39

Post by currymj » Tue, Jul 24, 2018, 10:38 AM UTC

furqs wrote: post #10

I have a question. Please excuse my ignorance but I thought that once the world has a working quantum computer, the world as we know it will be destroyed. Since a quantum computer can solve any NP hard problem in a polynomial time, it would mean it could break any kind of crypto, any kind of security and can brute force anything. Why hasn't that happened yet since its 2018 and we already have quantum computers?

First, they only work on really specific problems. They aren’t just a magic tool for brute force.

Current public-key crypto (both RSA and elliptic curve) happens to be one of those problems. However, there are systems where we don’t know how to break them with quantum computers, and it probably isn’t possible. These aren’t in wide use but have been tested in production e.g. by Google. If it becomes a problem, people can switch.

Second, actual existing quantum computers are too small to do much of anything. We are just hitting the point where they could start to become interesting. There are still engineering and theoretical challenges in making them really work.

All these quantum programming languages let you simulate a quantum computer, but doing so demands exponentially more resources as you add qubits. The advantage of a real quantum computer is that this would not be the case.

Re: Getting Started with Quantum Computing in Python

#40

Post by randomsearch » Tue, Jul 24, 2018, 12:24 PM UTC

throwaway37585 wrote: post #38

randomsearch wrote: post #37

Earlier quoted context omitted.

I mean, they do not illustrate when qubits have become entangled. You can't see that. I'm sure we can do better.

I’m not sure what you mean by “illustrate when qubits have become entangled”. Can you explain?

Entanglement is fundamental to QC, right? So in order to be a useful visualisation of an algorithm, a pictorial representation like a circuit should give some intuition to aid understanding such as: which qubits can potentially be entangled at a given stage in the circuit, etc.