Les Observables - iQuHACK2024

This repository contains the code and documentation for the project developed by Les Observablées during the MIT Quantum Hackathon (iQuHACK).

Our project focuses on addressing the Classiq challenge and leverages quantum computing techniques to find the most accurate approximation of tanh(x) in the domain of

f(x) : (0, 1] -> (0, 1]

We successfully employed a piecewise linear approximation to approximate tanh(x) within the domain (0,1]. In Stage 1, we achieved a precision of 7 with a max_distance of approximately 0.037, utilizing only 29 qubits. Moving on to Stage 2, we enhanced the precision to 31 while employing 98 qubits, achieving a circuit depth of just 1431. To gain a deeper understanding of the challenge, please refer to the document titled "Classiq’s Generalized Arithmetic Challenge.pdf."

Main Results:

Honourable Mention for the most creative algorithm

Algorithm approximation results (precision = 7)

Using 2-segment linear approximation and only one control bit, the approximation points plotted against $tanh(x)$

Theoretical approximation

Here you can see all the curves we tried throughout the hackathon

Team Members:

Aakash Warke

Chu-Hsuan (Abraham) Lin

Erica Sturm

Freesia Gaul

Yara Rodrigues

Installation

To install you will need a classiq account.

You will run:

!pip install -U classiq

import classiq
classiq.authenticate()

from classiq import *

Click on the hyperlink to validate your access