Feyorsh / Quantum_Bogosort

Bogosort, but with more shenanigans

About

This was my project that went from QED 2021 to the Chicago STEM fair and finally all the way to (virtual) ISEF. Everything is in the paper and you might even want to take a look at the slides, but if you have questions you can email me at [e-mail]. There's also a video presentation from city fair if you want to hear me stutter through a mediocre explanation of quantum mechanics. (The video on the ISEF website is pretty bad, if you actually want to learn more about the project I suggest this video.)

Wait... what the heck is quantum bogosort?

Long story short, I decided to take a joke algorithm (originally proposed in 2009) and actually implement it for a quantum computer. At it's core, I suppose one could call it an exercise in quantum software development and algorithmic design.

How is this different from other Quantum Random Number Generators (QRNGs)?

The main algorithim I developed for quantum bogosort is basically a QRNG, which creates $n \in \mathbb{Z}$ states in the range $[0, n)$, big endian; each of those states has amplitude $\frac{1}{\sqrt{n}}$ (and therefore $\frac{1}{n}$ chance of being measured). Every QRNG I've seen only works with $n$ values that are powers of $2$, because they just apply Hadamard to each qubit once and call it a day.

My algorithm, on the other hand, uses a bunch of cool stuff like arbitrary rotation gates, control gates, and 'divide-and-conquer' strategies to actually achieve a novel result (our friend $n$ can be any positive integer). Of course, for purposes of strictly RNG, my method is significantly more computationally complex and therefore less feasible. It's not meant to be useful or efficient.

How do you know this works?

I didn't do any statistical analysis of my findings, because I didn't have to; quantum circuits are just matrices. My algorithm can be mathematically verified as producing the correct output (although I don't know if this is the most efficient way to generate the correct output).

Where can I learn more?

1. Read the paper and/or slides.
2. Qiskit has some good educational material, in addition to being a really solid SDK (used in this project).
3. My personal recommendation for a starting textbook would be Chuang & Nielson's Quantum Computation and Quantum Information.

Setup

Note that you'll probably need to mess around with the code in testing.ipynb, especially if you're trying to get LaTeX to behave on your machine (I personally just used Overleaf). I used a really hacky implementation of qasm2circ and consequently my actual project structure looked a bit different from what you see.

Installation

Docker:

docker build -t <IMAGE_NAME> https://github.com/Borris-the-real-OG/Quantum_Bogosort.git
docker run --name <CONTAINER_NAME> -it <IMAGE_NAME>

Better yet, use a VSCode Dev Environment by using Remote-Containers: Clone Repository in Container Volume and plugging in this repo. If you already have a container up and running, you can attach to it with Remote-Containers: Attach to Running Container.

Docker? Humbug!

If you want to go at it the old-fashioned way, just make sure to install qiskit[visualization] for that quantum goodness.