juanli16/merminghzgame

Both classical (deterministic + Machine Learning) and quantum implementation of optimal strategies for the quantum pseudotelepathy game: Mermin-GHZ game

Mermin-GHZ Game

A comparison of classical and quantum strategies.

Quick pitch

The Mermin-GHZ game is an example of a simple game played by three independent actors. Classically players can only win a maximum of 3/4 times. However quantum players are capable of winning every time. While the applications of winning such a simple game are small, its an important expercise in testing the capabilities of modern quantum computers. In addition, since such systems are imperfect, it will become important to develop strategies to mitigate noise which can be tested on simple verifiable games such as this.

Implementation

We've developed our own random and optimal classical strategies to compare with a quantum algorithm we implemented in IBM's Qiskit quantum computing and simulation package. In addition, we presented all the results in Plotly's web-based Dash interface which makes it accessible to wide audience without the need for installing specialized packages. This kind of implementation is vital to expose younger students to quantum computing as the barrier to entry is currently too high for a generation not yet exposed to lots of programming and/or physics.

Submitted to the 2019 McGill Hackathon by Justin Li and Xavier Capaldi